32nd International Conference on Organization and Technology of Maintenance (OTO 2023) (Lecture Notes in Networks and Systems, 866) [1st ed. 2024] 3031514939, 9783031514937

Table of contents :
32nd International Scientific Conference Organization and Maintenance Technology OTO2023 (Osijek – Dec.12th, 2023)
Contents
About the Editors
Analysis of the Impact of Smartphone on the Environment Using the LCA Method
1 Introduction
2 Materials and Methods
2.1 Ecological Desing
2.2 Recycling
2.3 LCA – Life Cycle Assessment
2.4 Eco-Indicators
3 Example of the Application of the Eco-Indicator 99 Method
3.1 Developing a Life Cycle Tree Using SimaPro 8.0.5
3.2 Results of Product Life Cycle Assessment (LCA)
3.3 Product Impact on the Environment by Catogory
3.4 Normalization of Results
3.5 Scoring of Results
4 Conclusion
References
Lean Smart Maintenance for Machine Tools
1 Introduction
2 Experimental Approach
2.1 Creation of an Evaluation Matrix for Maintenance
2.2 Selection of a Suitable Maintenance Software Landscape
2.3 Target Processes for the Maintenance Software
3 Conclusion and Outlook
References
Advanced Construction Materials Based on Concrete to Protect the Living Space from Non-Ionizing Radiation
1 Introduction
2 Protection of Nonionizing Radiation – Shielding
3 Measurement of EM Wave Propagation Through Concrete
4 Conclusion
References
Analysis of the Structure of Agricultural Machinery Repair Workshops - A Case Study
1 Introduction
2 Materials and Methods
3 Results and Discussion
4 Conclusion
References
Maintenance of Agricultural Machinery in the Company Jerković d.o.o.
1 Introduction
2 Materials and Methods
3 Results and Discussion
3.1 Claas Axion 960
3.2 Claas Axion 830
3.3 Claas Arion 530
3.4 Claas Arion 430
3.5 Case CS 105 Pro
3.6 Claas Lexion 6900 and Claas Trion 650
4 Conclusion
References
Bridging the Physical and Virtual Worlds: A Hand Tracking Gesture Recognition System for XR Applications
1 Introduction
2 Overview of the Field of Gesture Recognition in Augmented Reality Technologies
3 Technological Tools and Platforms for the Experiment
3.1 Unity Game Engine
3.2 Oculus Quest 2
3.3 Oculus SDK
3.4 OctoXR
4 Gesture Detection Algorithm
5 Algorithm Implementation and Application
5.1 Rock, Scissors, Paper
5.2 Gesture Hero
6 Conclusion
References
Transmission of Electromagnetic Waves Through a Clay Material
1 Introduction
2 Electromagnetic Parameters of the Shield Material
3 Simulation Calculation of Coupling Parameters Through a Clay Block
4 Conclusion
References
Opening Doors and Drawers by a UR5 Robot with Force Control
1 Introduction
1.1 Related Research
2 Trajectory
2.1 Door Opening Trajectory
2.2 Drawer Opening Trajectory
3 Position/Force Control
4 Experimental Evaluation
5 Conclusion
References
Maintaining Mobile Communication in Distress and Emergency Situations
1 Introduction
2 Radio Amateurs in Distress and Emergency Situations
3 Recent Experiences, Earthquake in Petrinja
4 Radio Amateur Handheld Radio Stations
4.1 Antennas for Handheld Radio Stations
4.2 Directional Antennas
5 An Amateur Radio Repeater
5.1 Simplex Repeater
6 Maintaining Communication in Urban Conditions
6.1 Scenario A
6.2 Scenario B
6.3 Scenario C
7 Results Analysis
8 Conclusion
References
Testing the Quality of CNC Plasma Thermal Cutting in Accordance with the HRN EN 1090-2 Standard for the Production of Steel Structures
1 Introduction
2 Testing the Quality of Thermal Cutting According to the Requirements of HRN EN 1090-2
2.1 Measuring the Verticality of the Cutting Surface
2.2 Measurement of the Roughness of the Cutting Surface
2.3 Measuring the Hardness of the Cutting Surface
3 Conclusion
References
Automated Titration of SO2 in the Winery Environment: Conceptual Design and Proof of Concept
1 Introduction
2 Titration of SO2 in Wine
2.1 Overview of Methods for SO2 Detection and Measurement in Liquidous Compounds
2.2 SO2 Detection and Measurement in Wine and Wine Derivates
3 Automated Titration of SO2 in Wine Production
3.1 Titration Hardware Concept and Design
3.2 System Software and System Automation – An Algorithm and State Machine
4 Proof of Concept and Experimental Validation
4.1 Testing and Proofing Methodology
4.2 Results and Analysis
4.3 Implementation of Neural Network
5 Discussion and Conclusion
References
RS485 Network Design and Maintenance in Food Processing Industry: A Winery Application
1 Introduction
2 Communication in Processes Environment
2.1 Communication Challenges and Topologies Used in Processes
2.2 EMI Hardened Differential Pair-Based Serial Communication - RS485
3 RS485 Network in Winery Applications
3.1 Winery Environment Challenges and Requirements for Process Control
3.2 RS485 Topology for Multi-Nodal Smart Wine Parameters Measurement
4 RS485 Winery Network Analysis: An Experimental and Functional Analysis
4.1 Testing Methodology and Analysis Objectives
4.2 Functional Analysis and Experimental Proving of Topology Concept
5 Discussion and Conclusion
References
IoT in Smart Chromodynamic Plants Gardening
1 Introduction
1.1 Paper Aim, Structure and Organization
2 Chromodynamic Plants Gardening
2.1 Plants Gardening – Principles and Challenges
2.2 Chromodynamic Supported Gardening
3 IoT Monitoring and Control in Chromodynamic Plants Gardening – Smart Gardening
3.1 IoT Hardware for Smart Gardening System
3.2 Software Support for System Monitoring and Control
4 Experimental Results and Analysis
4.1 Testing Methodology and Analysis Objectives
4.2 Results and Analysis
5 Discussion and Conclusions
References
Unmanned Aerial Vehicle Mapping of River Flow for Water Resources Management
1 Introduction
1.1 LiDAR Cameras in River Flow Mapping
2 Economic and Legal Aspects Related to River Flow Mapping
3 LiDAR (Light Detection and Ranging) System
3.1 LiDAR-Derived Data Used for River Flow Mapping
4 Conclusion
References
Development of a Device for Maintaining the Temperature of the Tendons During the Period of Recovery
1 Introduction
2 Design Phase
3 Device for Tendon Recovery
4 Findings and Research Analysis
References
Maintenance of Automobiles and Motorcycles Through Prism of OBD II Diagnostic Tools
1 Introduction
2 OBD II Connector Pinout and Protocols
3 OBD II Devices
3.1 ELM 327
3.2 K+DCAN/K+CAN, K-line
3.3 Stand-Alone Devices Based on the Standard OBD-II Protocol
3.4 Stand-Alone Devices Based on the Android Systems
3.5 Motorcycle Related Diagnostic Tools
4 Application of OBD II Diagnostic Device in the Field of Maintenance
5 Conclusion
References
Construction of Variable Sheet Metal Hand Bending Tool
1 Introduction
1.1 Basics of Sheet Metal Bending
2 Materials
3 Tool Construction
4 FEM Numerical Simulation
5 Conclusion
References
Cost-Effectiveness of an Automatic Lubrication System for Bearings
1 Introduction
1.1 Bearings
1.2 Functions of Lubrication in Rolling Bearings
1.3 The Influence of Lubrication on Industrial Systems
1.4 Lubrication Film Analysis in Oil Lubrication
1.5 Automatic Lubrication Systems and SmartLub
2 Material and Method
2.1 Economic Contribution of Automatic Lubrication Systems
2.2 Calculation of Oil Requirements
3 Results Discussion
4 Conclusion
References
Combining DOE and EDAS Methods for Multi-criteria Decision Making
1 Introduction
2 EDAS Method
3 Recommended Method
4 Applied Cases
4.1 Case 1
4.2 Case 2
4.3 Case 3
5 Conclusion
References
Testing the Durability of the Color of Façade Materials
1 Introduction
1.1 General Information
1.2 Previous Works Dedicated to Color Durability of Building Materials
2 Materials and Methods
2.1 Materials
2.2 Methods
3 Results and Discussion
4 Conclusion
References
Dimensional Measuring System with Temperature Compensation
1 Introduction
2 Description of the Measurement System and Operation
3 Description of the Measurement Program for Measuring with Temperature Compensation
4 Workflow and Presentation of Achieved
5 Conclusion
.References
3D Printing Technology: Materials, Application and Current Trends in Process Improvement
1 Introduction
2 Materials Used for 3D Printing
3 3D Printing Applications
4 3D Printing Improvement Trends
5 Conclusion
References
Characteristics, Manufacturing, and Testing Methods of Polymer Gears: Review
1 Introduction
2 Test Methods and Key Characteristics
3 Polymer Materials in Gears Manufacturing
4 Manufacturing of Polymer Gears
5 Conclusion
References
Energy Efficiency Enhancement in a Small Industrial Facility
1 Introduction
2 Reactive Power Compensation in Distribution Network
2.1 Traditional Methods
2.2 Modern Methods
3 Small Industrial Facility
4 Results
5 Discussion
6 Conclusion
References
Comparative Study of Single-Input and Dual-Input PSS in Multi-machine System
1 Introduction
2 Eigenvalues and Participation Analysis
3 Case Study
4 Discussion
5 Conclusion
References
Creation and Maintenance of Public Real Estate Records - Business Models in Croatia
1 Introduction
2 Croatian Real Estate Registry System
2.1 Land Registry Reform
2.2 Cadastral Surveys
3 Situation in the Real Estate Registry in the Republic of Croatia
3.1 Insufficient Synchronization of Real Estate Records
3.2 How to Arrange Real Estate
4 Geodetic Activities
4.1 Organization of Geodetic Activities
4.2 State and Circumstances of Previous Years
4.3 Current Situation (2023)
5 Economic Position of Geodetic Activity
5.1 Tenders for Land Surveying Jobs
5.2 Market Framework of Geodetic Works in the Republic of Croatia
5.3 Economic Position of Employees of Surveying Companies
6 Business Model of the Geodetic Company
6.1 Business Expenses of the Geodetic Company
6.2 Employee Motivation
6.3 How to Run a Business in Unfavorable Conditions?
7 Concluding Remarks
References
Preventive Maintenance of Hydraulic Press Using Magnetic Testing
1 Introduction
2 Magnetic Tests
3 Advantages and Limitations of Magnetic Tests
3.1 Magnetizing Currents
4 Testing Procedure on the Hydraulic Press
5 Conclusion
References
Diagnostics of Journal Fluid-Film Bearing Failure Using a Data Manager System – Case History
1 Introduction
2 Technical Data of the Gear Unit
3 Monitoring System and Data Manager System
4 Diagnostics of Journal Fluid-Film Bearing Failure
5 Fact Finding After Disassembling the Machine
6 Conclusions
References
Predicting the Specific Gravity of Must During Fermentation Using Machine Learning Models
1 Introduction
2 Methodology in Related Research
3 Wine Fermentation Data
4 Fermentation Modelling
4.1 Evaluation of the Models
4.2 Machine Learning Models
5 Conclusions
References
Importance of Blackbody in Everyday Infrared Thermography
1 A Short Historical Introduction to Infrared Thermography
2 Basic Properties of Thermal Radiation
3 Infrared Thermal Cameras and Blackbody
3.1 Infrared Thermal Cameras
3.2 Black Body
4 Calibration of Infrared Thermal Camera
4.1 Calibration of Infrared Thermal Camera with Blackbody
4.2 Calibration of Infrared Thermal Camera with the Reference Objects
5 Conclusion
References
The 5S Method and Its Strategic Determinants Within the Organization of Production Plants
1 Introduction
2 Strategic Determinants and the Concept of the 5S Method
3 Implementation of the 5S Method
3.1 Preparatory Steps for the Implementation of the 5S Method
3.2 Stages of Application of 5S Tools
4 Usefulness and Possible Results of Applying the 5S Method in the Production Plants
4.1 Prerequisites for Successful Implementation of the 5S Method in Production Facilities
4.2 Development and Directions of the 5S Method
5 Conclusion
References
The Impact of Electric Car Charging on the Power Grid
1 Introduction
2 Measurement Analysis
3 Conclusion
References
Investigation of the Improvement of the Wheat Endosperm Hardness Assessment Method by Improved Construction of the Grain Cutting Knife
1 Introduction
2 Materials and Methods
3 Results and Discussion
4 Conclusions
References
Computer Network Design for Office Building Environment
1 Introduction
2 Technical Description of the Office Building’s Computer Network
2.1 Structured Cabling
2.2 Equipment Marking System
2.3 Marking of the Network Distributor
2.4 Examination of the Installed Passive Network
2.5 Vertical Network, Logical Diagram of a Computer Network
2.6 Horizontal Network, Listing the Connectors and Cables of the Building’s Computer Network, and Network Equipment
3 Quality of Service (QoS)
3.1 Layer 2, and Layer 3 Packet Marking
3.2 Management and Avoidance of Network Congestion, Rejection and Reduction of Network Traffic
4 Conclusion
References
An Advanced Tactile-Haptic Controller for Smart Home
1 Introduction
1.1 Paper Structure and Organization
2 Tactile-Haptic Systems in Smart Homes – An Overview and Theory of Operation
2.1 Tactile-Haptic Systems in UI Interaction
2.2 Tactile-Haptic Controllers Based on IoT for Smart Homes
3 Tactile-Haptic IoT Based Interactive Controller
3.1 Design of the Hardware – Structural and Functional Design
3.2 Software Solution for UI Interaction and Smart Home Deep Integration
4 System Test and Validation
4.1 Test and Validation Methodology
4.2 Results and Analysis of the Test and Validation
5 Discussion and Conclusion
References
Improving Maintenance Planning with the Help of Information Technologies
1 Introduction
2 Processing of Maintenance Work Orders in the MP2 Professional with Cost Analysis
2.1 Annual Maintenance Activities at the Selected Equipment
2.2 Creating Work Orders for Annual Maintenance of the Selected Equipment
3 Purchasing Process with the MP2 Professional
4 Conclusion
References
Author Index

Citation preview

Lecture Notes in Networks and Systems 866

Tomislav Keser Naida Ademović Eleonora Desnica Ivan Grgić   Editors

32nd International Conference on Organization and Technology of Maintenance (OTO 2023)

Lecture Notes in Networks and Systems

866

Series Editor Janusz Kacprzyk , Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland

Advisory Editors Fernando Gomide, Department of Computer Engineering and Automation—DCA, School of Electrical and Computer Engineering—FEEC, University of Campinas— UNICAMP, São Paulo, Brazil Okyay Kaynak, Department of Electrical and Electronic Engineering, Bogazici University, Istanbul, Türkiye Derong Liu, Department of Electrical and Computer Engineering, University of Illinois at Chicago, Chicago, USA Institute of Automation, Chinese Academy of Sciences, Beijing, China Witold Pedrycz, Department of Electrical and Computer Engineering, University of Alberta, Alberta, Canada Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland Marios M. Polycarpou, Department of Electrical and Computer Engineering, KIOS Research Center for Intelligent Systems and Networks, University of Cyprus, Nicosia, Cyprus Imre J. Rudas, Óbuda University, Budapest, Hungary Jun Wang, Department of Computer Science, City University of Hong Kong, Kowloon, Hong Kong

The series “Lecture Notes in Networks and Systems” publishes the latest developments in Networks and Systems—quickly, informally and with high quality. Original research reported in proceedings and post-proceedings represents the core of LNNS. Volumes published in LNNS embrace all aspects and subfields of, as well as new challenges in, Networks and Systems. The series contains proceedings and edited volumes in systems and networks, spanning the areas of Cyber-Physical Systems, Autonomous Systems, Sensor Networks, Control Systems, Energy Systems, Automotive Systems, Biological Systems, Vehicular Networking and Connected Vehicles, Aerospace Systems, Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power Systems, Robotics, Social Systems, Economic Systems and other. Of particular value to both the contributors and the readership are the short publication timeframe and the worldwide distribution and exposure which enable both a wide and rapid dissemination of research output. The series covers the theory, applications, and perspectives on the state of the art and future developments relevant to systems and networks, decision making, control, complex processes and related areas, as embedded in the fields of interdisciplinary and applied sciences, engineering, computer science, physics, economics, social, and life sciences, as well as the paradigms and methodologies behind them. Indexed by SCOPUS, INSPEC, WTI Frankfurt eG, zbMATH, SCImago. All books published in the series are submitted for consideration in Web of Science. For proposals from Asia please contact Aninda Bose ([email protected]).

Tomislav Keser · Naida Ademovi´c · Eleonora Desnica · Ivan Grgi´c Editors

32nd International Conference on Organization and Technology of Maintenance (OTO 2023)

Editors Tomislav Keser Faculty of Electrical Engineering, Computer Science and Information Technology University of Osijek Osijek, Croatia Eleonora Desnica Technical Faculty “Mihajlo Pupin” University of Novi Sad Zrenjanin, Serbia

Naida Ademovi´c Faculty of Civil Engineering University of Sarajevo Sarajevo, Bosnia and Herzegovina Ivan Grgi´c Mechanical Engineering Faculty University of Slavonski Brod Slavonski Brod, Croatia

ISSN 2367-3370 ISSN 2367-3389 (electronic) Lecture Notes in Networks and Systems ISBN 978-3-031-51493-7 ISBN 978-3-031-51494-4 (eBook) https://doi.org/10.1007/978-3-031-51494-4 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

32nd International Scientific Conference Organization and Maintenance Technology OTO2023 (Osijek – Dec.12th, 2023)

Editorial Board Tomislav Keser (President) Naida Ademovi´c Eleonora Desnica Ivan Grgi´c

International Programme Committee Naida Ademovi´c, BiH Samir Avdakovi´c, BiH Marinko Barukˇci´c, Croatia Damir Blaževi´c, Croatia Dominika Crnjac-Mili´c, Croatia Josip Cumin, Croatia Eleonora Desnica, Serbia Marijana Hadzima-Nyarko, Croatia Tomáš Hanák, Czech Republic Lajos Jozsa, Hungary Hrvoje Glavaš (President), Croatia Mirko Karakaši´c, Croatia Tomislav Keser, Croatia Gyuyong Kim, South Korea Imre Kiss, Romania Stefanija Klaric, Australia ˇ Caslav Livada, Croatia Francisco Martínez-Álvarez, Spain Krešimir Nenadi´c, Croatia Ljiljana Radovanovi´c, Serbia Łukasz Sadowski, Poland Vasilija Sarac, Republic of North Macedonia Nataša Šuman, Slovenia Ismar Voli´c, USA

vi

32nd International Scientific Conference

Scientific Committee Naida Ademovi´c, BiH José-Lázaro Amaro-Mellado, Spain Ivan Ambroš, Croatia Aleksandar N. Ašonja, Serbia Samir Avdakovi´c, BiH Josip Balen, Croatia Tomislav Bari´c, Croatia Ðuro Banaj, Croatia Marinko Barukˇci´c, Croatia Huseyin Bilgin, Albania Damir Blaževi´c, Croatia Mirjana Bošnjak-Kleˇcina, Croatia Borko Bulaji´c, Serbia Dominika Crnjac-Mili´c, Croatia Nenad Cvetkovi´c, Serbia Slawomir Czarnecki, Poland Josip Cumin, Croatia Sanja Dimter, Croatia Zlata Dolaˇcek-Alduk, Croatia Tihomir Dokšanovi´c, Croatia - c, Serbia Snežana Ðurdi´ Mohamed Elchalakani, Australia Krešimir Fekete, Croatia Irena Gali´c, Croatia Ranko Gantner, Croatia Mario Gali´c, Croatia Goran Gazi´c, Croatia Hrvoje Glavaš, Croatia Ivan Grgi´c, Croatia Krešimir Grgi´c, Croatia Ehsan Harirchian, Germany Bassam A. Tayeh, Palestine Ivana Hartmann Toli´c, Croatia Mostafa Fahmi Hassanein, Egypt Željko Hocenski, Croatia Ercan I¸sık, Turkey Irena Ištoka Otkovi´c, Croatia Aleksandar Juri´c, Croatia Josip Job, Croatia Željka Jurkovi´c, Croatia Tanja Kalman Šipoš, Croatia Tomislav Keser (President), Croatia Zvonimir Klai´c, Croatia

32nd International Scientific Conference

Goran Kneževi´c, Croatia Veljko Kokovi´c, Serbia Pejo Konjati´c, Croatia Mirko Köhler, Croatia Ivan Kraus, Croatia Hrvoje Krsti´c, Croatia Danijel Kukaras, Serbia Krešimir Lackovi´c, Croatia ˇ Caslav Livada, Croatia Silva Lozanˇci´c, Croatia Ivica Luki´c, Croatia Siniša Mariˇci´c, Croatia Predrag Mari´c, Croatia Krešimir Mastanjevi´c, Croatia Kristina Mastanjevi´c, Croatia Antonio Morales Esteban, Spain Paweł Niewiadomski, Poland Srete Nikolovski, Croatia Emmanuel Karlo Nyarko, Croatia Svilen Radoslavov Raˇcev, Bulgaria Dorin Radu, Romania Mirsad Rašˇci´c, BiH Ivica Petrovi´c, Croatia Hugo Rodrigues, Portugal Goran Rozing, Croatia Sebastijan Seme, Slovenia Dina Stober, Croatia Marinko Stojkov, Croatia Ivana Šandrk Nuki´c, Croatia Marija Šiško Kuliš, Croatia Damir Šljivac, Croatia Marija Šperac, Croatia Željko Špoljari´c, Croatia Vedrana Jerkovi´c Štil, Croatia Andrej Štrukelj, Slovenia Lidija Tadi´c, Croatia Danijel Topi´c, Croatia Ismar Voli´c, USA Nikola Vukašinovi´c, Slovenia

vii

viii

32nd International Scientific Conference

Organizing Committee Damir Blaževi´c (President) Tomislav Bari´c Krešimir Mastanjevi´c Marinko Stojkov Ružica Kljaji´c, mag.ing.

Technical and IT Support Dario Došen, mag.ing. Davor Begi´c, bacc.ing. Mario Miloloža, mag.ing. Igor Sušenka, dipl.ing.

Contents

Analysis of the Impact of Smartphone on the Environment Using the LCA Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jure Mariji´c, Marko Vili´c, Ivan Grgi´c, Mirko Karakaši´c, Željko Ivandi´c, and Domagoj Komarˇci´c Lean Smart Maintenance for Machine Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Josip Florian Strutz, Tomislav Saric, Katica Simunovic, and Ivan Samardzic Advanced Construction Materials Based on Concrete to Protect the Living Space from Non-Ionizing Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vanja Mandri´c, Slavko Rupˇci´c, Davor Vinko, and Ivica Dominkovi´c Analysis of the Structure of Agricultural Machinery Repair Workshops A Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Željko Baraˇc, Tomislav Juri´c, Ivan Plašˇcak, Monika Markovi´c, ´ c Antonija Koji´c, and Denis Cosi´ Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o. . . . . . . . ´ c, Željko Baraˇc, Tomislav Juri´c, Ivan Plašˇcak, Denis Cosi´ and Josip Damjan Bridging the Physical and Virtual Worlds: A Hand Tracking Gesture Recognition System for XR Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ˇ Matija Fumi´c and Caslav Livada

1

14

25

35

42

53

Transmission of Electromagnetic Waves Through a Clay Material . . . . . . . . . . . . Vanja Mandri´c, Slavko Rupˇci´c, Davor Vinko, and Domagoj Bilandžija

75

Opening Doors and Drawers by a UR5 Robot with Force Control . . . . . . . . . . . . . Jana Duki´c, Lukrecia Vuli´c, Valentin Šimundi´c, Petra Peji´c, and Robert Cupec

86

Maintaining Mobile Communication in Distress and Emergency Situations . . . . Tomislav Bari´c and Hrvoje Glavaš

97

Testing the Quality of CNC Plasma Thermal Cutting in Accordance with the HRN EN 1090-2 Standard for the Production of Steel Structures . . . . . . 113 - Marija Stoi´c, Josip Cumin, Miroslav Duspara, Ivan Dunder, and Antun Stoi´c

x

Contents

Automated Titration of SO2 in the Winery Environment: Conceptual Design and Proof of Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Tomislav Keser, Robert Miling, Davorin Miliˇcevi´c, and Damir Blaževi´c RS485 Network Design and Maintenance in Food Processing Industry: A Winery Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Ivana Kovaˇcevi´c, Tomislav Mati´c, Tomislav Keser, and Robert Miling IoT in Smart Chromodynamic Plants Gardening . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Željko Juric, Tomislav Keser, Ivor Plander, and Mario Levani´c Unmanned Aerial Vehicle Mapping of River Flow for Water Resources Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Marina Peko, Dominika Crnjac Mili´c, and Ivan Vidakovi´c Development of a Device for Maintaining the Temperature of the Tendons During the Period of Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 Ivan Grgi´c, Mirko Karakaši´c, Željko Ivandi´c, Jure Mariji´c, and Marko Vili´c Maintenance of Automobiles and Motorcycles Through Prism of OBD II Diagnostic Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Josip Cumin, Daniel Novoselovi´c, Dejan Mari´c, and Tomislav Šoli´c Construction of Variable Sheet Metal Hand Bending Tool . . . . . . . . . . . . . . . . . . . 184 Josip Cumin, Hrvoje Vorel, Miroslav Duspara, and Hrvoje Glavaš Cost-Effectiveness of an Automatic Lubrication System for Bearings . . . . . . . . . 199 Serkan Yildiz, Murat Apakhan, and Muharrem Hilmi Aksoy Combining DOE and EDAS Methods for Multi-criteria Decision Making . . . . . . 210 Do Duc Trung, Nguyen Xuan Truong, Hoang Tien Dung, and Aleksandar Ašonja Testing the Durability of the Color of Façade Materials . . . . . . . . . . . . . . . . . . . . . 228 Piotr Kosi´nski and Agata Jodko Dimensional Measuring System with Temperature Compensation . . . . . . . . . . . . 241 Nemanja Zuji´c and Djordje Dihovicni 3D Printing Technology: Materials, Application and Current Trends in Process Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Ivan Palinkas, Eleonora Desnica, Jasmina Pekez, Aleksandar Rajic, and Milan Rackov

Contents

xi

Characteristics, Manufacturing, and Testing Methods of Polymer Gears: Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Ana Markovi´c, Lozica Ivanovi´c, and Blaža Stojanovi´c Energy Efficiency Enhancement in a Small Industrial Facility . . . . . . . . . . . . . . . . 283 Ružica Kljaji´c, Zorislav Kraus, Krešimir Fekete, and Predrag Mari´c Comparative Study of Single-Input and Dual-Input PSS in Multi-machine System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Tomislav Košorog, Muharem Mehmedovi´c, Predrag Mari´c, and Ružica Kljaji´c Creation and Maintenance of Public Real Estate Records - Business Models in Croatia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 Milan Ivanovi´c, Franjo Ambroš, and Vedran Stojnovi´c Preventive Maintenance of Hydraulic Press Using Magnetic Testing . . . . . . . . . . 330 Ljiljana Radovanovi´c, Borivoj Novakovi´c, Mi´ca Djurdjev, and Luka Djordjevi´c Diagnostics of Journal Fluid-Film Bearing Failure Using a Data Manager System – Case History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342 Marko Katini´c, Pejo Konjati´c, Mirko Karakaši´c, and Danko Glavaš Predicting the Specific Gravity of Must During Fermentation Using Machine Learning Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 Ivana Kovaˇcevi´c, Mihaela Ori´c, Ivana Hartmann Toli´c, and Emmanuel Karlo Nyarko Importance of Blackbody in Everyday Infrared Thermography . . . . . . . . . . . . . . . 364 Hrvoje Glavaš The 5S Method and Its Strategic Determinants Within the Organization of Production Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 Dominika Crnjac Mili´c The Impact of Electric Car Charging on the Power Grid . . . . . . . . . . . . . . . . . . . . . 386 Zvonimir Klai´c, Filip Ðakovi´c, Mario Primorac, and Krešimir Fekete Investigation of the Improvement of the Wheat Endosperm Hardness Assessment Method by Improved Construction of the Grain Cutting Knife . . . . . 394 Vinko Krstanovi´c, Kristina Habschied, and Krešimir Mastanjevi´c Computer Network Design for Office Building Environment . . . . . . . . . . . . . . . . . 402 Damir Blaževi´c, Tomislav Keser, and Matko Mance

xii

Contents

An Advanced Tactile-Haptic Controller for Smart Home . . . . . . . . . . . . . . . . . . . . 419 Miloš Radi´c, Tomislav Keser, and Damir Blaževi´c Improving Maintenance Planning with the Help of Information Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 Sanda Simunovic, Tomislav Saric, Andrijana Milinovic, and Iva Samardzic Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447

About the Editors

Tomislav Keser (1976) was born in Vinkovci, Republic of Croatia. He graduated from the Faculty of Electrical Engineering in Osijek in 2000 and received his PhD in computer engineering from the same faculty in 2009. Since graduation, he has been working at the Faculty of Electrical Engineering, Computer Science and Information Technology in Osijek, first as an assistant and now as an associate professor. His areas of interest for research and teaching are computer engineering, embedded computing, computer vision and system automation. In addition, he has recently distinguished himself in research and activity in computer engineering for space applications, especially embedded computer systems in LEO applications. He is a long-time member of IEEE and a member of several technical groups. He also maintains and leads activities and is chair of an IEEE Croatian Section on Systems, Man and Cybernetics. He is a member of the Croatian Institute for Standards in the Standards Section, where he is a board member of the Robotics and Automation Committee. He is currently an associate professor at the University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, where he lectures and teaches courses in computer science, automation and embedded computing technologies and leads and/or collaborates on numerous scientific projects. Naida Ademovi´c (1973) was born in Sarajevo, Bosnia and Herzegovina, and is Associate Professor at the University of Sarajevo, Faculty of Civil Engineering. She gained her degrees at the University of Sarajevo, Faculty of Civil Engineering, Bosnia and Herzegovina (Bachelor of Sciences and PhD), Rühr Universität, Fakultät für Bauingenieurwesen, in Bochum (Germany) (Master of Science); University of Padova (Italy) and University of Minho, Guimarães (Portugal) (Advanced Master of Science). Her scientific and professional activity covers the field of earthquake engineering, concrete and masonry structures, and bridges, finite element modeling and resilience of structures. She worked on numerous professional and scientific local and international projects. She is a member of scientific committees of international journals, and she was involved in organizing several conferences. As well, she is an active reviewer for several international journals. She worked on projects regarding Integrating Seismic Risk Considerations into Energy Efficiency Investments in the Western Balkans and Improving Sarajevo’s Resilience through Urban Regeneration. She has participated in several scientific projects and has been engaged in research in the field of concrete and masonry structures, earthquake engineering and earthquake risk assessment. She is an author/coauthor of five books. She participated in the establishment of an interdisciplinary study program Protection against Natural Disasters at the Center for Interdisciplinary Studies, University of Sarajevo, where she is Associate Professor. She is a member of the Ph.D. study course at the University of Padova (Italy). She has published more than 100 scientific and professional papers.

xiv

About the Editors

Eleonora Desnica (1971) was born in Zrenjanin, Serbia. She graduated at the University of Novi Sad, Faculty of Technical Sciences in Novi Sad, Mechanical Engineering. She achieved MSc (in 2004) and PhD degree in technical science in 2010 at Technical faculty “Mihajlo Pupin” Zrenjanin, University of Novi Sad. Since 1998, she has been employed at Technical Faculty, “Mihajlo Pupin” in Zrenjanin. She is a member of Department of Mechanical Engineering and became full professor in 2021. Her fields of work are as follows: computer-supported technologies in mechanical engineering and their application in education, CAD/CAM, machine design, methodology of engineering design, thermography in industrial practice, mechanics and machine mechanisms, and engineering graphic communications. In her professional and scientific work, she has published as author or coauthor over 100 papers in national and international journals and reports at scientific professional meetings. She is a member of scientific committees of national and international journals and organizing several conferences. She works as a researcher on two projects—one domestic and one international. Ivan Grgi´c (1985) in Slavonski Brod where he finished elementary and secondary technical school. In 2005, he enrolled at the Faculty of Mechanical Engineering in Slavonski Brod. He graduated in 2010 with his final work with the topic “Numerical model of heat transfer by conduction and convection in a waste incinerator” under the mentorship of Marija Živi´c, PhD, as one of the ten most successful students of his generation. In 2011, he enrolled as a part-time student in a postgraduate doctoral study, course “Product Design and Numerical Modelling”. In February 2020, he finished his doctoral thesis with the topic “Development, manufacturing and application of technical system for determination of biomechanical properties of gracilis and quadriceps muscle tendon”. Since February 2020, he has been employed as Postdoctoral Researcher at the Mechanical Engineering Faculty in Slavonski Brod, Department for Mechanical Design.

Analysis of the Impact of Smartphone on the Environment Using the LCA Method Jure Mariji´c(B)

, Marko Vili´c, Ivan Grgi´c , Mirko Karakaši´c, Željko Ivandi´c, and Domagoj Komarˇci´c

Mechanical Engineering Faculty in Slavonski Brod, University of Slavonski Brod, Trg I. B. Mažurani´c 2, 35000 Slavonski Brod, Croatia [email protected]

Abstract. Today’s industrial production has a great impact on the environment all over the world. The constant increase in the world’s population has a great impact on the expanding industries in the world. Maintenance, eco-design and life cycle analysis are important concepts related to sus-tainability and environmental protection. Life cycle refers to everything associated with the product at all stages of its life, from the extraction of raw materials to the disposal or recycling of the product. The mobile phone was chosen as a representative in accordance with the global development of the relevant industry and the desire to show its impact on the environment using the LCA (Life cycle assessment) method. The concepts of sustainable development, recyclability and eco-indicators were explained and some concepts and tools were presented to show how can mobile phones industry reduce the impact and con-tribute to the environment and sustainability. Keywords: eco-design · life cycle analysis · Eco-Indicator 99 · smartphone

1 Introduction Every product has an impact on the environment during its production, use and at the end of its life cycle. The life cycle refers to everything related to the product at all stages of its life, from the extraction of raw materials to the disposal or recycling of the product. It is necessary to extract the raw material, manufacture, package and distribute the finished product and dispose of it at the end of its life. If the use of the product involves the consumption of materials and/or energy, this phase of the product’s life cycle has a significant impact on the environment. In view of today’s trend towards large-scale and mass production, and with the emphasis on quality and economy of products, it is necessary to work continuously on the development of tools and equipment [1]. Products can be divided into short-term and durable goods. Short-term goods are products that are intended for one-time use; these are products such as food and beverages that must be consumed within a certain period of time after they are used. Durable goods are products that are intended to last longer than short-lived goods, and most often the manufacturers of such products are required to guarantee the product for a certain period © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 1–13, 2024. https://doi.org/10.1007/978-3-031-51494-4_1

2

J. Mariji´c et al.

of time. These are products such as household appliances, machines, cars, furniture, etc. [2]. As the product and technology, itself evolves for production, as well as the manufacturer, there is an increasing focus on an eco-approach with an environmentally friendly design and a choice of materials that can be reused for different purposes. However, waste gives an insight into the lives and behaviour of past generations and also reflects the image of the materials that humans have used. For example, materials that did not exist in past centuries have appeared in the waste of our time and have found use in all areas of contemporary life due to their many advantages - plastic and rubber [2]. Maintenance, ecological design and life cycle assessment are all important concepts related to sustainability and environmental protection. Maintenance refers to procedures and activities carried out to maintain the quality and functionality of a product or system during its lifetime. This may include regular repairs, replacement of parts, cleaning and maintenance to ensure that the product or system continues to perform its function without creating additional waste or undesirable environmental impacts. Ecodesign refers to the way in which products and systems are designed to minimise their negative impact on the environment throughout their life cycle. Life cycle assessment is a process that analyses the overall environmental impact of a product or system, from raw materials to waste disposal. It includes an assessment of all stages of production, use and disposal to identify the main factors that influence environmental impacts and to identify opportunities to improve sustainability. All three concepts are important to achieve sustainability and environmental protection. They are used to develop products and systems that are sustainable and have a minimal impact on the environment.

2 Materials and Methods 2.1 Ecological Desing Nowadays, designers, constructors and importers of products are expected to contribute to the reduction of energy consumption and pollution and to strive for energy efficiency in all phases of their activities. Ecological design as such means creating “smarter” products that contribute to environmental protection. All products have an impact on the environment during their life cycle, which includes all phases from “cradle to grave”. These stages include the use of raw materials and natural resources, production, packaging, transport, disposal and recycling. However, more than 80% of the environmental impact is determined in the design phase itself. Ecological design means that all impacts of products on the environment are considered at the earliest stages of design. This avoids, in particular, uncoordinated product design (of the type of removal of toxic substances, which should not be accompanied by increased energy consumption, which can also have a negative impact on the environment). The “Eco-Directive” provides a coherent and integrated framework that allows the establishment of mandatory requirements for some products in the field of ecological

Analysis of the Impact of Smartphone

3

design. For example, the “Standby Regulation’ requires that many electrical and electronic household appliances such as washing machines, televisions or PCs consume no more than 0.5 W in the so-called ‘off’ state since 2013.” [2]. However, the ecological design requirements must not compromise the functionality of the product or its safety, or have a negative impact on its availability and consumer health. The methodology has been developed to provide practical guidance to the Committee on how and when to assess which requirements and design approaches are appropriate for a particular product (Fig. 1).

Fig. 1. Extensive impact on the environment depending on investment and complexity

Requirements for ecological design: [3]. – are set separately for each individual product – set the minimum requirements for the performance of the product in order to reduce its impact on the environment – are mandatory for all products sold in the EU – are based on the environmental impact throughout the life cycle of the product (design, production, distribution and disposal). 2.2 Recycling Recycling or recovery, the process of processing waste materials and used products with the aim of obtaining raw materials and energy for reuse and use [4]. The complete waste management system combines waste streams, their collection and recovery with environmental benefits, economic optimisation and social acceptability. This system includes different materials (paper, plastic, glass, wood, metal, rubber), waste from different sources (households, industry) and different applications (packaging waste). In recycling, the aim is to use as much of the product as possible at the end of its life as a finished form or as a raw material to be returned to the production process

4

J. Mariji´c et al.

of the same or a new product. At the same time, the unusable part of the product must be reduced to the smallest possible size. Product-oriented production seeks to influence the environment through the most efficient, “ecological” product design possible and considers the impact on the environment throughout the life cycle. The emission of pollutants is monitored from the extraction of the raw material to the disposal of the used product. Some of the methods used are: – – – –

eco-design eco-efficiency LCA - life cycle assessment (product life cycle assessment) LCM - life cycle management.

According to [4], the circular economy system is based on the use of already used materials that can be renewed and used in new and different ways, which at the same time ensures a reduction in the exploitation of natural resources. Therefore, in 2015, the European Commission adopted a Circular Economy Package, which includes legislative proposals on waste, long-term targets to reduce disposal and increase recycling and reuse. The plan includes working on all parts of the chain, from production to reuse, and returning secondary raw materials and materials to the cycle [5]. 2.3 LCA – Life Cycle Assessment The life cycle of a product includes everything associated with the product in all phases of its life, i.e. from the extraction of raw materials to disposal or recycling. The methodology of life cycle assessment has the task of evaluating all components and environmental impacts associated with the product, process or activities during their lifetime [1]. The 5 Steps of a Product Life Cycle which can been seen on Fig. 2: [6]. 1. 2. 3. 4. 5.

Raw Material Extraction Manufacturing & Processing Transportation Usage & Retail Waste Disposal

LCA is a systematic approach consisting of four important steps which can been seen on Fig. 3: [6]. 1. Definition of the objective and scope - defining and describing products, processes and activities, determining the context and boundaries in which the assessment will be carried out. 2. Life Cycle Inventory (LCI) - identification and quantification of energy, water and materials consumed and pollution generation (air emissions, waste generation, …) 3. Life Cycle Impact Assessment (LCIA) - assessment of the potential ecological impact on the environment 4. Interpretation (explanation) of results - evaluation of results and impacts to select products, processes and services, clearly considering the uncertainties and predictions that affect the results.

Analysis of the Impact of Smartphone

5

Fig. 2. The 5 Steps of a Product Life Cycle

Fig. 3. Four key steps of the LCA method

LCA analyses of some products show that they have a significantly greater impact on the environment during use than during production. Various methods are used to assess the environmental impact of products. The best known and most widely used is the Eco-indicator 99, which is merely an extension of the LCA method, but not its simplification. LCA is by far the most reliable method for calculating environmental impact, but it also has its limitations. 2.4 Eco-Indicators Eco-indicators can be defined as numbers that express the overall impact of a product or process on the environment. These figures are derived from life cycle assessment data. The values of the eco-indicators are “considered” as dimensionless quantities, but the term point is used by common consent. It is also important to mention the scale of values, which was chosen so that 1 Pt (point) corresponds to one thousandth of the annual environmental impact of an average European inhabitant. The idea of an eco-indicator for a product’s impact on the environment was developed with the intention of promoting awareness of the totality of the material and living world, of which humans are a biological part. Incorporating the environment into the product design process is complex and long-term in every respect. Methods and programmes have been developed to simplify the process itself. The Eco-Indicator 99 has been the most widely used, as it shows the link between the impact of substances and harm to human health [2]. In order to calculate the Eco-indicator score, three steps are needed: [7].

6

J. Mariji´c et al.

1. Inventory of all relevant emissions, resource extractions and land-use in all processe that form the life cycle of a product. This is a standard procedure in Life Cycle Assessment (LCA) 2. Calculation of the damages these flows case to Human Health, Ecosystem Quality and Resources 3. Weighting of these three damage categories Three steps are illustrated (Fig 4.).

Fig. 4. General procedure for the calculation of Eco-indicator

It is important to emphasize that the disassembling procedure was done by the assessment and the free will of the authors. No disassembling procedure was followed or owned by the phone manufacturer. Similar approach was carried out in [8]. It is necessary to emphasize that the work focuses solely on the Eco Indicator 99 method. The Eco-Indicator 99 method was developed as a further development of the EcoIndicator 95 method. Based on the Eco-Indicator 95 method, improvements have been developed that allow a more accurate analysis of the environmental impact of a product or system. These include the updating of parameters, the introduction of new concepts such as toxicology indicator points (TIP) to assess toxicity, weighting factors that reflect current environmental priorities, and other improvements that have enabled a more comprehensive and accurate analysis of environmental impacts. The Eco-Indicator 99 method has become a useful tool for understanding and reducing the environmental footprint of a product or system and has surpassed its predecessor in its ability to assess environmental impact [9].

3 Example of the Application of the Eco-Indicator 99 Method The smartphone iPhone 4 was chosen as an example for the application of the Ecoindicator 99 method. The selected representative is chosen on the basis of its availability. The device is classified as electronic waste and it is recommended not to classify it as other waste. The device is assumed to be used several hours a day and its estimated lifetime is 4 years. It is assumed that it will then be recycled as electronic waste. For the analysis, it was necessary to separate all the parts of the device, then determine their mass and define the material. The mass was determined with a digital scale with an accuracy of 0.1 g. The parts of the smartphone can be seen in Fig. 5. Smartphone parts with a certain mass and material can be found in Table 1.

Analysis of the Impact of Smartphone

7

Fig. 5. Parts of the device Apple iPhone 4

Table 1. iPhone 4 mobile phone parts with masses and materials Name of the part

Mass [g]

Material

Pieces

Front housing (screen)

34,9

Glass

1

Central frame

37,2

Aluminum

1

Back cover

22,1

Glass

1

Battery

26,9

Lithium

1

Motherboard

13,9

Silicon

1

Speaker housing

2,5

PVC

1

Camera

1,2

Aluminum + Glass

1

Protective sheets (total mass of all)

1,1

Aluminum

1

SIM card slot

0,7

Aluminum

1

Vibration

0,7

Aluminum

1

3.5 mm headphone jack

0,4

PVC

1

However, some parts could not be weighed with a digital scale, such as screws. All the screws together have a mass of less than 0.1 g, so these parts are ignored.

8

J. Mariji´c et al.

Once all the necessary data had been obtained, it was entered into the SimaPro 8.0.5 program [10], the desired eco-indicator was selected and the analysis started. The results obtained are presented in tables and diagrams. 3.1 Developing a Life Cycle Tree Using SimaPro 8.0.5 In developing the product life cycle tree, which shows the materials used in the product, the electricity consumed by the product, the production processes used to make the parts and the type of waste disposal, it was necessary to enter the type of material and the mass for each part. The upper part of the life cycle tree of the appliance can been sean in Fig. 6.

Fig. 6. Life cycle tree with eco-indicator values

3.2 Results of Product Life Cycle Assessment (LCA) As a result of the assessment of the life cycle of the product, a table was obtained in which the following are listed: – – – – – –

all substances contained in the materials of the device substances required for the production of raw materials substances used in the manufacture of parts all forms of energy used in all phases of the life cycle of the product radioactive radiation from certain substances occupies the earth’s surface Part of the results of the LCA of the appliance in Fig. 7.

Analysis of the Impact of Smartphone

9

Fig. 7. The results of the life cycle assessment of the appliance

3.3 Product Impact on the Environment by Catogory The impact of devices on three basic categories - human health, ecosystem quality and resources - is shown in Fig. 8.

Fig. 8. The environmental impact of device in three basic categories

The results are also presented according to the effects in more precisely defined categories. This definition includes: carcinogenic effects, respiratory effects, climate change, radiation effects, ozone layer effects, ecotoxicity, eutrophication, depletion of the earth’s surface and fossil fuel consumption effects. The above results can be seen in Fig. 9.

10

J. Mariji´c et al.

Fig. 9. The environmental impact of device

In the diagrams, the individual categories are scaled with 100%. The blue colour shows the percentages indicating the impact of materials that are stored, i.e. disposed of in landfill. In addition to considering the entire assembly, the impact of individual materials and production processes can also be considered, as shown in Fig. 10.

Fig. 10. The environmental impact of the materials (used for the appliance)

3.4 Normalization of Results Since in the previous diagrams all categories are scaled to 100%, it is difficult to determine which material or technological process related to the product has the greatest impact on the environment. To get a better representation of the product’s impact on the environment, it was necessary to perform normalisation. This is an evaluation procedure that makes the individual impacts comparable. Figure 11 shows the normalised results in a diagram.

Analysis of the Impact of Smartphone

11

Fig. 11. Normalized results in a diagram

3.5 Scoring of Results Relatively comparable results obtained by normalisation do not show the importance of individual influences, so that the results obtained so far for two different products were not comparable. By applying weighting factors to the normalised results, we obtain diagrams in which the environmental impacts are expressed in points (Pt). 1 Pt represents one thousandth of the environmental impact of an average European during one year. The following figures show the impact of the device with the corresponding points. Figure 12. Shows the assessed categories and the assessed results, while Fig. 13. Shows the assessed materials by specific categories.

Fig. 12. Assessed categories and the assessed results

12

J. Mariji´c et al.

Fig. 13. Impact of materials on the environment - scored results

4 Conclusion The multitude of different demands placed on production or other units are the main reason for the negative impact on the environment. Nevertheless, proper consideration and investigation of the existing problem provide greater opportunities for a comprehensive and proper analysis of the problem that needs to be solved. A detailed analysis using the SimaPro software can be used to easily determine the results of the environmental impact of the mobile devices mentioned above. Every product has a greater or lesser impact on the environment and it is of great importance to consider the materials from which it is made, the possibilities of recycling and the impact of production on the environment. According to the life cycle tree, it is obvious that silver has the greatest impact on the environment. Therefore, recycling should be a solution for the disposal of those types of compounds that may affect the human habitat in some way. The final result of the whole life cycle is 2.67 pt. Since this result includes the whole process, from the impact of each material, the electricity consumption, the way each part is produced, to the disposal in the designated places, it can be concluded that the device does not have an excessive impact on the environment. From the diagrams obtained, it is possible to identify the shortcomings of the design and technical solutions of the existing product. Recycling of the entire product would reduce the impact on the environment as opposed to the assumed disposal in household waste. From the point of view of suitability for material recycling, the number of different types of materials is minimised. The incompatibility of materials is a kind of measure of the imperfection of the available recycling processes. In conclusion, it is very difficult, if not impossible, to answer the question of which criteria are most important and which should be the focus of the whole process. Acknowledgment. We would like to thank SimaPro company for granting the programme used to conduct the research mentioned in this article.

Analysis of the Impact of Smartphone

13

References 1. Mariji´c, J., Vili´c, M., Grgi´c, I., Karakaši´c, M., Ivandi´c, Ž.: Development, structure and design od stamping tool. In: Glavaš, H., Hadzima-Nyarko, M., Karakaši´c, M., Ademovi´c, N., Avdakovi´c, S. (eds.) 30th International Conference on Organization and Technology of Maintenance (OTO 2021). OTO 2021. LNNS, vol. 369, pp. 50–70. Springer, Cham (2022). https://doi.org/10.1007/978-3-030-92851-3_4 2. Kljajin, M., Opali´c, M., Pintari´c, A.: Recycling of electrical and electronic products. Faculty of Mechanical Engineering in Slavonski Brod, Slavonski Brod (2006) 3. Your Europe Homepage. https://europa.eu/youreurope/business/product-requirements/com pliance/ecodesign/index_hr.htm. Accessed 28 Apr 2023 4. Croatian encyclopedia Homepage. https://enciklopedija.hr/Natuknica.aspx?ID=52144. Accessed 28 Apr 2023 5. Vrbek, M.: Ecological Product Design: The Foundation of Circular Waste Management. Master’s thesis. University of Zagreb, Faculty of Economics and Business, Zagrab (2020). https://urn.nsk.hr/urn:nbn:hr:148:410383 6. Ecochain Homepage. https://ecochain.com/knowledge/life-cycle-assessment-lca-guide/. Accessed 28 Apr 2023 7. Goedkoop, M., Spriensma, R.: Eco-indicator 99 Manual for Designers, A damage-oriented method for Life Cycle Impact Assessment. Ministry of housing, Spatial Planning and the Environment, The Hague (2000) 8. Vukašinovi´c, J., Grgi´c, I., Mariji´c, J., Karakaši´c, M.: Reverse engineering and 3D printing of the spinning reel: from maintenance to the new product design. In: Blaževi´c, D., Ademovi´c, N., Bari´c, T., Cumin, J., Desnica, E. (eds.) 31st International Conference on Organization and Technology of Maintenance (OTO 2022). OTO 2022. LNNS, vol. 592, pp. 106–117. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-21429-5_10 9. The Eco-indicator 95 Final Repoer Homepage. https://pre-sustainability.com/legacy/dow nload/EI95FinalReport.pdf. Accessed 04 Sep 2023 10. SimaPro 8.0.5 For Education Homepage. https://simapro.com/. Accessed 28 Apr 2023

Lean Smart Maintenance for Machine Tools Josip Florian Strutz1,2(B) , Tomislav Saric2 , Katica Simunovic2 and Ivan Samardzic2

,

1 Ceratizit Besigheim GmbH, 74354 Besigheim, Germany

[email protected] 2 Mechanical Engineering Faculty in Slavonski Brod, University of Slavonski Brod, Slavonski

Brod, Croatia

Abstract. The Lean Smart Maintenance concept has been developed to provide an optimal and targeted maintenance strategy. By implementing this concept, manufacturing companies can achieve an efficient maintenance program for their CNC machines, which is a crucial component in ensuring the continued operation and the overall success of the manufacturing process. Moreover, maintenance resources are minimized as maintenance measures are performed more effectively. Both planned and unplanned shutdowns are considered. In this paper, the focus is on a comprehensive consideration of maintenance in a SAP-based, mediumsized company with an inhomogeneous machine park and the digital networking of all departments. The data-based model follows a centralized approach from the maintenance technician’s mobile device, over the warehouse management of the maintenance to sales information with the use of advanced planning and scheduling software. Keywords: Lean Smart Maintenance · Machine tool · Digitalization · SAP EAM · Asset management

1 Introduction In this paper, a concept for the maintenance of machine tools according to the principles of lean smart maintenance (LSM) is presented. The aim is to increase the long-term availability of the machines by performing the maintenance measures in a condition-oriented and predictive manner using modern, data-based methods [1]. The overall system presented here is not limited to data-based methods, but also includes failure-oriented and preventive maintenance as required. With the correct application of all three maintenance strategies and the support of a software-based control system, a plant evaluation and classification should be possible. The LSM idea describes a complete management system, which covers the goal of a high reliability and availability of plant components, but also the loss minimizing maintenance execution [1]. In the long term, the aim is to add value to the entire company. In addition, maintenance resources should be used efficiently to reduce costs, minimize downtime and improve plannability. The implementation of these goals is illustrated using an example company. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 14–24, 2024. https://doi.org/10.1007/978-3-031-51494-4_2

Lean Smart Maintenance for Machine Tools

15

Increasing demands for flexibility, reliability and speed in manufacturing are accompanied by an increased need for automation and complexity. This poses great challenges for maintenance. Nowadays, maintenance rarely focuses on the entire plant, but preferably on individual components [2, 3]. This makes sense not only from the point of view of minimizing downtime, but also from the point of view of steadily increasing specialization of maintenance staff. It is generally accepted that maintenance is becoming increasingly important. A comprehensive consideration of business and technical aspects is of decisive importance for effective maintenance management. An important tool in this context is maintenance planning. This requires an optimal mix of reactive, preventive and status-oriented maintenance strategies. These strategies must be individually evaluated and dynamically adapted to success [4, 5]. Furthermore, it is essential for successful companies to plan production according to plant availability. This requires production planning strategies with low uncertainty. Internal and external planning uncertainties are known to every supply chain and provide new challenges and the balancing of safety and risk on a daily basis. These challenges must be reduced through internal uncertainties in maintenance. One of these is essentially the availability planning of machine tools. In an ideal production planning of machine resources, the trade-off of planning reliability does not depend on the individual supply chain staff, but on data-based models that have a bidirectional interface - the digital twin of the maintenance process [6]. Since digital twins are extremely complex to model and their development does not always pay off in a reasonable time, alternatives are often used in reality. For example, automatic adaptation of the model is usually dispensed with and only easily evaluable parameters are used. In mass production, this can be the number of pieces; in single-item and small batch production, it is necessary to fall back on runtime-dependent data. In addition, there are also errors in the modeling that must be minimized. These include the input variables of the model, residual variations (same set variables - different results), measurement errors, interpolation errors in the model, parametric variations, etc. [7].

2 Experimental Approach The aim of this paper is to develop a maintenance model that not only describes the maintenance processes, but also enables an overall view of maintenance and servicing. The holistic processes of maintenance and production planning are taken into account, resulting in a practical maintenance model. In the initial situation of the company under consideration, SAP enterprise asset management (EAM) is already used in maintenance. This system already offers several advantages, which will be explained in this section. Further on, a time recording of several maintenance staff has to be carried out in order to objectively identify the painpoints. Since the optimization potentials are to be recorded and processed in a structured manner, a survey of the maintenance employees has to be carried out in order to gain a subjective impression of the initial situation. With the software SAP EAM already in use, an excellent basis has already been created. The use of this extremely flexible and comprehensive software module from SAP enables central integration into the existing SAP ERP system (Fig. 1). This effectively

16

J. F. Strutz et al.

prevents redundant data, as there are no interfaces at which redundancies are created. In addition, no information is lost, as data is stored exclusively in one database, making interfaces obsolete.

Fig. 1. Centrally organized data structure of the ERP system with the subordinate subareas of the SAP EAM software suit.

The structured recording and processing of fault messages is also possible. Among other aspects, this has the advantage of being able to explicitly assign the required resources to each operation. The creation can be carried out by any user, which avoids latencies in the failure demand creation. The necessary transparency of the order status can also be controlled by every user, since the information is stored centrally in the ERP system. In this paper, three different maintenance strategies are considered: • Failure-oriented maintenance strategy [8] • Preventive maintenance strategy [9, 10] • Condition-oriented maintenance strategy [11]. The maintenance strategy currently consists only of strategies one and two. Failureoriented maintenance reacts exclusively to component failures, i.e. repairs. It reacts only after the wear reserve of a component has been used up. In order to avoid unplanned and high downtimes of equipment, preventive maintenance focuses on high equipment availability. Maintenance work on a machine tool is carried out at a predefined interval. In order to nevertheless cover a large part of the wear and tear, it is necessary to determine as precisely as possible the time of failure of the respective unit under consideration. Therefore, the knowledge about the failure behavior, the load as well as the utilization time and reliability of the unit is very important for an optimal maintenance time [9].

Lean Smart Maintenance for Machine Tools

17

2.1 Creation of an Evaluation Matrix for Maintenance The evaluation matrix (Fig. 2) serves as the basis of the SLM process and is intended to define the risk cost potential and the maintenance strategy. It provides an important basis for determining the maintenance strategy for specific machines and assemblies.

Fig. 2. Evaluation matrix of an LSM process to determine prioritization rank and maintenance strategy.

The maintenance objects are examined for their failure consequences and their impact on the company - but also for the maintenance costs incurred. From the causes of downtime, the plant priority can be determined (rank A to C). The health, safety & environment (HSE) status is listed here as a representative of all concerns that are not foreseeable but affect at least one of these HSE areas. A possible example for a machine tool is a leakage of the cooling lubricant. A hose could be defective, causing cooling lubricant to leak and enter the groundwater. In this case, immediate maintenance intervention is required, so no further maintenance strategies are considered. The same applies to any error that leads to machine downtime. These usually cause an on-time delivery (OTD) problem, requiring immediate response. Another problem with unplanned machine downtime is, among other things, redundant operators who should be occupied with other activities.

18

J. F. Strutz et al.

While these two concerns are immediately given a high priority, the root cause must be investigated after the failure has been corrected. Ideally, the cause of downtime can be covered by condition-based maintenance in the future, allowing rank C to be defined. This ensures optimal plannability for the maintenance staff, but also for the production. Service life is a very simple type of preventive maintenance. However, reasons for preventive maintenance measures may not only be necessary from the point of view of the risk of failure, they may also be necessitated by legal requirements. For example, in the case of machine tools, the cyclical inspection of jaw chucks and pressure accumulators ensure fixed maintenance intervals that may not be extended. However, if the risk of failure and thus the costs are considered high, the maintenance interval must be shortened. The cost analysis includes all cumulative costs related to the failure - machine hourly rate, employee costs for production and maintenance, spare parts, lubricants, etc. The cost threshold is defined here as the fourth quartile (>Q3) of all machine repairs of similar machines. This ensures that high-precision grinding machines are not compared to milling machines for rough machining. If the repair costs are in the last quartile, a short-term preventive strategy is recommended. Another reason for preventive action is to maintain a fixed grooving time. In the case of machine tools, for example, this category includes checking for compressed air leaks, as these are cost-intensive and can be responsible for possible malfunctions and cycle time extensions. Since these can be planned, they are assigned priority rank C. In addition, however, it is possible that there is a predictive maintenance (PdM) model for this reason. The last category includes the criteria operating time and frequency. The operating time is not to be mixed up with the service life, because the operating time only refers to the time in which the component is used. In the case of hydraulics, for example, a distinction can be made between the state machine switched on and machining process. As a rule, the hydraulics are used, among other things, for clamping the workpieces. It is irrelevant whether chips are currently being produced, as the workpiece must also remain clamped for the setup process. Consequently, the hydraulic system operates for a longer period of time than chips are produced. The frequency (number of pieces, number of strokes,…) is also a main reason for the consumption of the wear reserve. To stay with the example of the clamping device, it plays a significant role how often the clamping cylinder is moved until the seals are worn out and have to be replaced. In the PdM model, however, it is not only the tribological wear of the seals that must be considered, but also their aging. These causes fall into prioritization rank C, since they are predictable - analogous to rank B. However, a shift does not ensure legal consequences this is the reason for this lower ranking. However, they also ensure perfect planning of production and maintenance. PdMs for machine tools planned for implementation are as follows: 1. 2. 3. 4. 5.

Central cooling lubricant supply with condition sensors, particles and filling level Central hydraulics with condition sensors & filling level Pressure monitoring of pressure accumulators in hydraulics Monitoring of compressed air leakages at the machines Differential pressure of exhaust filters of the cooling lubricant mist.

Lean Smart Maintenance for Machine Tools

19

The focus of the solutions discussed here is primarily on economic considerations in the sense of the lean concept. Recently, many companies and publications have been dealing with various possibilities of PdM models (often using machine learning). However, upon closer examination, not all of them are economically feasible, especially if they are developed in-house. An example of common publications is the use of tool life models in machining, such as in [12], or vibration measurement of spindle bearings and ball screws. These solutions are worthwhile in mass production where one type of machine is used redundantly, but not in medium sized companies - as it is considered here. Medium-sized companies are usually characterized by a high product diversity and small to medium-sized quantities, which is why they often have an inhomogeneous machine park. Consequently, the model has to be applied anew for each machine, since the mechanical surfaces are different. 2.2 Selection of a Suitable Maintenance Software Landscape With the determination of the SLM evaluation matrix and the PdMs, it is possible to specify the requirements for the software landscape in detail. The focus here is on seamless integration of the acquired data on the store floor, their evaluation and the derivation of maintenance measures. In the infrastructure layer, a Beckhoff industrial PC is used as an edge device (ED) and connected to all machine controllers. The advantage of such an integration is that it is possible with all controllers and thus a factory standard is defined. The ED runs parallel to the PLC and only picks up the programmed signals. There is no signal processing with the PLC or the actual machine control, which means that performance and safety of the running machine are not affected. Additional sensors, which are not necessary for the machine operation, are integrated directly at the ED in order to forward the data. This reduces integration costs and avoids warranty issues with machine manufacturers. Data-based processes for maintenance require a rethinking of previous system landscapes. On the one hand, this involves the introduction of software that can handle large volumes of data. SAP databases are certainly suitable, but it is only since the introduction of the S4/Hana version that SAP has offered detailed evaluation options for this data. In this case, the project of integrating ED’s started already before the S4 update. In addition, the SAP cost structure is not optimal for data volumes such as those generated by high-frequency process analysis. With Microsoft’s Azure solutions, customized packages are available that offer a better cost structure. These new analysis methods, require the breaking of the typical layer structures of a company, see Fig. 3. As shown in the figure, a new two-part data layer has been added and the reporting layer has also been extended and divided into two parts. The data layer contains the necessary Azure packages, but they form a closed overall system. This means that the data scientist still has very few points of contact with the SAP process world and can focus fully on data analysis and PdM model creation. Nevertheless, for a closed loop it is necessary to create an interface to the ERP, which is made possible with SAP business warehouse (BW). For example, to generate SAP EAM orders automatically.

20

J. F. Strutz et al.

Fig. 3. Software landscape with the different layers and new data layer for the LSM.

A new part has also been added to the reporting layer, as a distinction must be made here between the users of the processes and the official reporting. The official reporting is e.g. characterized by the reporting to the management and other official instances. The new part on the right side of the reporting layer is designed for data science and the dynamic and volatile daily business. Here, the user can create dashboards and trigger predefined actions or receive notifications on the smartphone. When using maintenance software, it is essential to eliminate all documentation in Excel or other non-system data formats. For example, machine documentation must also be available in digital form in order to access the data on the machine to be repaired. Filing maintenance plans and repair cards is still a common medium, but is contrary to end-toend digitalization. The integration of mobile devices is indispensable to ensure that these processes can be carried out by the maintenance staff throughout and directly at the plant (core layer). Thus, [13] shows a significant increase in productivity in different areas of maintenance through mobile devices. Subsequent maintenance and repair documentation or the requesting of spare parts can also be carried out directly. In typical maintenance, a continuous warehouse management is often omitted. The reasons for this are manifold, but previous time recordings clearly show the necessity of warehouse management. Since the SAP infrastructure already exists in the company, the SAP Extended Warehouse Management (EWM) extension is a logical step (core layer, Fig. 3).

Lean Smart Maintenance for Machine Tools

21

2.3 Target Processes for the Maintenance Software The LSM target processes are based in their entirety on the results shown in Fig. 2. The various maintenance strategies ultimately always generate a request in SAP EAM, whereby the priority field is filled with the corresponding rank. This field is used in detailed maintenance planning as an evaluation criterion for the processing sequence. Inventory management is often dispensed with in maintenance. A decisive reason for this are the employees who are not bound to a particular location. Central warehouse management in the maintenance or logistics buildings is only useful to a limited extent, since very long distances are involved, especially for repairs. The reason for this is that the tools and spare parts required are not known in advance. Centralizing the warehouse in logistics would prevent redundancies, but slow down the process for the maintenance engineer. An efficient solution while accepting redundancies is described in the following text. The manufacturing company already has several warehouse lifts on the shop floor, but they do not have SAP EWM connectivity. Therefore, the connection of the warehouse lifts to the warehouse management system is carried out completely, also for the components of the production. If components are removed from the lifts, they must be booked out in SAP in advance, and only then is the tray made available. This process efficiently prevents stock shortages. Furthermore, access to these trays is only possible for maintenance staff. Figure 4 shows the overall process flow, which is explained below. Unforeseen and thus manually created EAM demands (e.g. machine downtimes) are an exception and must therefore be evaluated according to on-time delivery (OTD). For this purpose, the failure must first be viewed in order to estimate the downtime. This requires communication between the two detailed planning tools from production and from maintenance. The ranking is limited to A and B (Fig. 2) only, because there are two unknowns at this stage: the maintenance duration and the delivery time of the spare parts, since these are not known in advance. A rigid ranking on rank A to evaluate the maintenance duration is not used, because this can shift already existing orders intraday. Furthermore, it should be mentioned that the separation of the detailed planning tools is necessary, since in production a daily update at 0:00 is sufficient. In the detailed planning of maintenance, this short update time is used to be able to react intraday to machine failures. Thus, in the detailed planning of the maintenance there is an update in real time. As a result, maintenance tasks that have already been started may no longer be given the highest priority due to the planning run at 0:00 h. For example, a repair of an A-ranking machine started today can be suspended the next day because the machine is downgraded in the ranking by the planning run of the production at 0:00 o’clock due to the standstill of a more OTD-critical machine. This circumstance provides for a longer repair duration, by setup times with the maintenance engineer, but ensures a very good OTD and brings thus a high customer satisfaction. However, this approach can of course be adapted to the company’s requirements. At the end of an unplanned shutdown or manually created demand, the sales department may be informed about the postponed delivery date for the customer. Planned shutdowns, on the other hand, are predictable. This means that the higher delivery times for new production orders are already taken into account in the quotation

22

J. F. Strutz et al.

Fig. 4. Process diagram for different requirements and disturbances

process. Since maintenance is a repetitive activity, there are predefined maintenance packages. The planned downtimes therefore automatically ensure a request in SAP EWM at the nearest warehouse lift of the machine as soon as the requirement is created. If the spare part is not available, it is ordered automatically through the ERP and taken into account in both detailed planning tools. Here, too, an automatic delivery date shift takes place in Sales. The rest of the process is analogous to the one described above, but without the prior assessment of the damage.

Lean Smart Maintenance for Machine Tools

23

3 Conclusion and Outlook In summary, this paper presents how machine tools are to be handled in a modern, data-based maintenance. The focus is on the process landscape and the development of intelligent maintenance models. The environment in the described company consists of an inhomogeneous machine park and low production quantities, which is common in medium-sized companies. Furthermore, it must be mentioned that the abstract listing of the prioritization presented here does not replace an FMEA or FMECA analysis for each machine. However, the above-mentioned supercategories provide an indication of the possible defect groups of each machine tool. For an overall maintenance concept, each machine and its components must be examined in detail for the probability of failure and its consequences for other components and production. A supplement to the process methodology illustrated here is offered, among other things, by reliability centered maintenance (RCM), since these topics are examined in detail on the basis of seven questions. A possible further development option is the implementation of prescriptive maintenance. In this case, the intelligent models of PdM are not only used to predict errors and failures, but also to actively suggest preventive measures. The avoidance measures can then be implemented manually or automatically in the manufacturing process. The conditions for automatic implementation are already possible independently of the control system thanks to the installed edge device.

References 1. Biedermann, H., Kinz, A.: Lean smart maintenance-value adding, flexible, and intelligent asset management. BHM Berg- Huettenmaenn. Monatsh.g- Huettenmaenn. Monatsh. 164, 13–18 (2019) 2. Bokrantz, J., Skoogh, A., Berlin, C., Wuest, T., Stahre, J.: Smart maintenance: a research agenda for industrial maintenance management. Int. J. Prod. Econ. 224, 107547 (2020) 3. Liebstückel, K.: Instandhaltung mit SAP S/4HANA: Das Praxishandbuch, 6th edn. Rheinwerk Verlag, Bonn (2023) 4. Mostafa, S., Lee, S.-H., Dumrak, J., Chileshe, N., Soltan, H.: Lean thinking for a maintenance process. Prod. Manuf. Res. 3, 236–272 (2015) 5. Eckpunkte und Ausgestaltung eines Fremdfirmencontrollings. Der Instandhaltungs-Berater. Vollmüller B (2018) 6. Wittmeir, T., Oettl, F., Schilp, J.: Digitaler Zwilling für die additive Fertigung/Digital twin for additive manufacturing. WT Werkstattechnik 113(3), 119–123 (2023) 7. Denkena, B., Wichmann, M., Kettelmann, S.: Prozesskettenplanung unter Unsicherheit/Process chain planning under uncertainty - consideration of internal and external uncertainties during planning of process chains across companies. WT Werkstattechnik 113(4), 146–152 (2023) 8. Schuh, G., Lorenz, B.: TPM – eine Basis für die wertorientierte Instandhaltung. In: Reichel, J., Müller, G., Mandelartz, J. (eds.) Betriebliche Instandhaltung. VDI-Buch, pp. 79–81. Springer, Berlin, Heidelberg (2009). https://doi.org/10.1007/978-3-642-00502-2_6 9. Reitz, A.: Lean TPM: 12 Schritten zum schlanken Managementsystem; effektive Prozesse für alle Unternehmensbereiche; gesteigerte Wettbewerbsfähigkeit durch KVP; Erfolge messen mit der Lean-TPM-Scorecard. 3. unveränderte Auflage. Mi-Fachverlag, München (2014)

24

J. F. Strutz et al.

ˇ 10. Coruši´ c, J.: Unapredenja održavanja tunelskih ventilatora lean managementom/Improvement of maintenance of tunnel fans with lean management. Master Thesis. Strojarski fakultet u Slavonskom Brodu, Slavonski Brod, mentor: Šari´c, T. (2020) 11. Rötzel, A., Rötzel-Schwunk, I.: Instandhaltung: Eine betriebliche Herausforderung. 5. überarbeitete und erweiterte Auflage. VDE VERLAG GmbH, Berlin, Offenbach (2017) 12. Chen, J.-Y., Lin, Y.-L., Lee, B.-Y.: Development of the adaptive system for tool management. Tehnicki vjesnik-Technical Gazette 30(2), 648–654 (2023) 13. Mobile EAM & PM solution with form engine and consumer grade UX. On Device Solutions Ltd., Birmingham

Advanced Construction Materials Based on Concrete to Protect the Living Space from Non-Ionizing Radiation Vanja Mandri´c(B)

, Slavko Rupˇci´c , Davor Vinko , and Ivica Dominkovi´c

Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, Croatia [email protected]

Abstract. This paper presents measurements of the reflection and transmission coefficient of electromagnetic waves through concrete and two concrete-based composites: concrete with steel fibers and concrete with carbon fibers with the aim of enhancing the attenuation of electromagnetic waves passing through concrete and concrete composites. The frequency range for which the measurements were carried out extends from 30 MHz to 18 GHz and includes most of the existing stationary (and mobile) radiation EM sources in the environment in which the general population moves (FM, DVBT2, 2G - 5G mobile radio systems, stationary radar border area control systems). The measurement results show that as the thickness of the concrete increases, the transmission through concrete and concrete composites decreases (parameter S12 decreases). Furthermore, composites with carbon fibers partially reduce, while those with steel fibers significantly reduce, the transmission of EM waves through such composites. The reduction of parameter S12 by composites with steel is up to 52 dB (at a frequency of 2.9 GHz) for a block thickness of 100 mm. It is important to emphasize that in the case of composites with steel fibers, the influence of the fibers on the transmission parameters is more significant than the thickness of the sample. The reason for this is the increase in electrical conductivity of composites with steel fibers due to the increased proportion of conductive components in the concrete. Keywords: Concrete – Based Composites · EM Wave Attenuation · EM Wave Propagation · Non - Ionizing Radiation · Transmission Parameter · 2G – 5G Systems

1 Introduction The present work deals with the transmission and reflection of electromagnetic waves passing through concrete blocks and concrete-based composites in the frequency range of 30 MHz–18 GHz. Many scientists and researchers have been involved in the development of the model and the measurements of the coupling parameters, among which I would like to highlight: U.B. Halabe and K. Maser, who in 1989 developed models for the attenuation © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 25–34, 2024. https://doi.org/10.1007/978-3-031-51494-4_3

26

V. Mandri´c et al.

of electromagnetic waves in concrete as a function of frequency, temperature, moisture content, chloride content and constituents of the concrete mix [1]; in 1997 W.C. Stone studied the propagation of electromagnetic waves in bricks, concrete, glass and wood by measuring the attenuation, electrical conductivity and dielectric constant of the material [2]; in 2008 Y. Pinhasi created a model that takes into account multiple reflections through the space where both the transmitter and receiver are located, including wall, ceiling and floor reflections [3]; in 2012, the American National Institute of Standards and Technology (NIST) and the German Armed Forces University conducted research on building materials from the point of view of protection against electromagnetic fields in the field of mobile communications 3G, 4G and LTE, as well as digital television frequencies, GPS and wireless smart metres, testing the following: concrete, brick, wood, drywall, plywood, glass and reinforcement [4]. M.R. Mahfouz, et al. 2014 measured RF attenuation in different types of walls: dry walls (up to 70 GHz), concrete (up to 3 GHz) and bricks (up to 10 GHz) [5]. ITU-R P.2040-1/2015 provides guidelines on the influence of the electrical properties of building materials and structures on the propagation of radio waves above 100 MHz and provides basic equations for the calculation of reflection and transmission coefficients when EM waves pass through building materials [6]. J. Kurunvilla et al. 2019 deals with materials of natural origin, thermoset and thermoplastic polymers, metal embedded matrices, biodegradable polymers, nanomaterials and textile materials, cement-based materials as well as possible fillers for these materials (carbon, metals, polymers) and their possible applications [7]. Their parameters and damping properties that can potentially be used for EM protection. The energy of EM waves can be absorbed when passing through concrete, and this absorption is influenced by: material parameters, chemical and mechanical construction, moisture, age and the amount of conductive and dissipative particles. This absorption is manifested by attenuation of the EM wave. The key electrical parameters of damping concrete are dielectric constant and electrical conductivity. In 2018, Zhekov measured the attenuation of EM waves propagating through solid concrete block walls in the frequency range from 400 MHz to 2.7 GHz [8]. The second chapter of the paper deals with protection of nonionizing radiation – electromagnetic shielding. The third chapter deals with the measurement of EM wave propagation. The last chapter covers all the relevant conclusions of this paper as well as the possibility of continuing the research of EM wave transmission and reflection through concrete and other building materials in the future.

2 Protection of Nonionizing Radiation – Shielding Electromagnetic shielding is one of the oldest, most widespread and effective methods of protection against the effects of EM radiation (Fig. 1). It is based on shielding those objects (subjects), devices or spaces that need to be protected from EM radiation. In this work, it is assumed that the armoring of a residential space where people live, and the armoring is made as a concrete (concrete-based composite) shield that reduces the effect of EM radiation on people inside that space (object). The effectiveness of such

Advanced Construction Materials

27

armoring can be expressed by the armoring effectiveness factor which can be calculated from the electric field ratio Eq. (1) measured without armor (unshield - ush) and with armor (shield - sh), at the same point of observation: SEe = 20 log

Eush , dB Esh

(1)

The shielding effectiveness can be more simply determined by the ratio of transmission parameters S21 without and with protection as: SES21 = 20log

S21ush S21sh

(2)

The most common way to determine SE is from the measured results of transmission parameters S21 (or S12 ). In such a measurement, in addition to the transmission parameters, the reflection parameters (S11 and S22 ) are also measured, which give us information about the part of the energy of the EM wave that is reflected on the armor (Fig. 1). A material that has high reflection and low transmission is a material that provides high protection against EM radiation.

Fig. 1. Transmission and reflection of EM wave when passing through a concrete (or concrete based composite) wall

Observed from the point of view of the impact of EM energy on the environment, the best quality protective material would be the one with the highest absorption of EM energy, with minimal energy reflection into the environment.

3 Measurement of EM Wave Propagation Through Concrete Measurements of coupling parameters were performed using Vector Network Analyzer Anritsu MS2038C and two types of antennas: conical dipole antenna - BicoLOG30100, AARONIA and log-periodic antenna - HyperLOG 30200, AARONIA (Fig. 1, Fig. 2 and Fig. 3). In this work, measurements of the transmission parameters of S21 were carried out on three different concrete samples: a) pure concrete with a density of about 1600 kg/m3 , dimensions - width × height: 600 × 600 and three thicknesses: 100, 200 and 300 mm; b) concrete-based composite with the addition of carbon fibers (CF) in the amount of 1.5% by mass; c) concrete-based

28

V. Mandri´c et al.

Fig. 2. Schematic representation of the system for measuring coupling parameters (transmission and scattering) using two types of antenna: BicoLog30100, AARONINA, 30 MHz–1 GHz and HyperLog 30200, AARONIA, 380 MHz–19 GHz.

Fig. 3. Measuring antennas when measuring coupling coefficients (transmission and reflection) using two types of antennas: a) HyperLOG 30200, AARONIA – log-periodic antenna; b) BicoLOG30100, AARONIA – conical dipole antenna.

composite with the addition of steel fibers (SF) in the amount of 43 kg/m3 of concrete, dimensions - width × height: 600 × 600 and three thicknesses: 100, 200 and 300 mm. The measurements were carried out using two types of antennas: a conical dipole antenna for measurement in the frequency range from 30 MHz to 1 GHz, and a logarithmic antenna for the frequency range from 1 GHz to 18 GHz. Measurements of coupling parameters through blocks made of pure concrete show (Figs. 4, Fig. 5 - a) and b)), in both measured frequency ranges, a decrease in transmission through the blocks with increasing block thickness. It is important to note that the differences in the coupling parameter S12 with increasing block thickness are particularly prominent for the lower measured range and for part of the higher measured range (from 1–9 GHz). By increasing the frequency (it is especially visible in the range above 1 GHz) the transmission decreases. These are well-known conclusions that have been confirmed by these measurements. As far as the reflection parameters are concerned,

Advanced Construction Materials

29

there is no significant difference in the results between blocks of different thickness (from 100–300 mm). Nevertheless, it is evident that the values of the coupling parameters of the thicker blocks (200 and 300 mm) are much closer (and significantly lower) than the 100 mm thick block.

Fig. 4. Measurements results of coupling parameters of concrete in the frequency range 30 MHz to 1 GHz: a) S11 ; b) S12 .

Fig. 5. Measurements results of coupling parameters of concrete in the frequency range 1 to 18 GHz: a) S11 ; b) S12 .

Furthermore, the coupling parameters of two composites of the same thickness (200 mm) with different fibers were measured: carbon fibers and steel fibers (Fig. 6 and 7). For carbon fibers, there is a limiting factor that is conditioned by the reduction of the hardness of the concrete composite, and a composite with 1.5% mass fraction of carbon fibers was chosen, where there is no significant reduction in the hardness of the composites. In the case of composites with steel fibers, this limitation does not exist, as this additive further increases the hardness of the composite, but it can also affect the viscosity and the tensile and compressive forces that the composite withstands. Therefore,

30

V. Mandri´c et al.

a composite with a proportion of 43 kg of steel fibers per m3 of concrete was chosen. The comparison of these two materials from the point of view of their influence on EM coupling showed significantly better characteristics of the composite made with steel fibers (Fig. 6 and 7). For the range from 30 MHz to 1 GHz, the composite with steel fibers has an average of 15 dB lower values of the transmission parameter S12 , while for the range from 1 GHz to 18 GHz, this value of the difference is about 18 dB up to a frequency of 4 GHz (after that frequency the differences are not significant). As far as the reflection parameter is concerned, the differences are not significant, except that in almost the entire higher range, the reflection of the composite with carbon fibers is somewhat lower.

Fig. 6. Measurements results of coupling parameters of concrete, concrete composite with steel (SF43) and carbon (CF1.5) fibers in the frequency range 30 MHz to 1 GHz: a) S11 ; b) S12 .

Fig. 7. Measurements results of coupling parameters of concrete, concrete composite with steel (SF43) and carbon (CF1.5) fibers in the frequency range 1 to 18 GHz: a) S11 ; b) S12 .

The better material is the one that has a low value of the transmission parameter S12 and a high value of the reflection parameter S11 , which in this case is a composite with steel fibers. This is understandable because the introduction of electrically conductive

Advanced Construction Materials

31

material into the concrete structure (steel fiber) additionally increases the electrical conductivity of the composite. However, from the point of view of absorption of EM energy, a material that has low transmission and low reflection has a significantly higher absorption of EM energy. These results, therefore, can be viewed from these two rather different aspects. Composite with carbon fibers has higher transmission and low reflection than composite with steel fibers, so it does not fall into this category. This comparison of these two composites (along with the reference pure concrete) directed further research and measurement, which in the following focused only on composites with steel fibers. In the following, composites with steel fibers of different thicknesses are compared in more detail with reference blocks made of pure concrete (Fig. 8, Fig. 9 and Fig. 10).

Fig. 8. Measurement results of the transmission parameters of concrete-based composite with steel fibers (SF43) in the frequency range: a) 30 MHz to 1 GHz: b) 1 to 18 GHz.

Since the values of the reflection parameters differ slightly in the further results, the values of these parameters will not be listed, but only the parameters. If the composites with steel fibers are compared according to the S12 transmission parameters, then it is evident that for all three thicknesses of the block there is no significant difference in the value of the S12 parameter in the lower frequency range (Fig. 8 a)). A similar conclusion can be drawn for the higher frequency range. (Fig. 8 b)). In conclusion, it can be stated that for the lower measurement range, the thickness of the composite block with steel fibers (SF43) has no influence on the S12 parameters, and thus has no significant influence on the transmission of the EM wave through such a composite. Since, unlike composites with steel fibers, in the case of blocks made with pure concrete for the lower measurement range, the thickness of the block has a significant influence on the S12 parameters and thus a significant influence on the EM wave transmission through the concrete. This is visible in Fig. 9 and 10 for both measured frequency ranges. Thickness of composite blocks with steel fibers (SF43) and pure concrete in Fig. 9 and 10 is identical, and takes the values 100, 200 and 300 mm.

32

V. Mandri´c et al.

Fig. 9. Measurement results of the transmission parameters of concrete, concrete composite with steel (SF43) fibers in the frequency range 30 MHz to 1 GHz of thickness: a) 100 mm; b) 200 mm and c) 300 mm.

For the lower frequency range, the difference between the curves of the transmission parameters of S12 concrete and composites based on concrete with steel fibers is large for the smallest block thickness (100 mm), and the smallest for the largest block thickness (300 mm). The biggest difference between the S12 parameters of the concrete and SF43 composite with a thickness of 100 mm is −33 dB at the frequencies of 0.76 and 0.98 GHz, while the biggest difference of the blocks with a thickness of 300 mm is −22 dB at the frequency of 0.74 GHz. Namely, this does not mean that the 300 mm thick block is the worst in terms of S12 transmission parameter values, but that in composite blocks with SF43, the thickness does not affect the transmission as much as the addition of concrete - steel fibers. For the higher frequency range, the difference between the curves of transmission parameters of S12 concrete and composites based on concrete with steel fibers is large for the smallest block thickness (100 mm), and the smallest for the largest block thickness (300 mm) for different frequency sub-bands. Namely, by increasing the thickness of the block, the area in which this difference is visible decreases. For a thickness of 100 mm, this subband is from 1 to 9.5 GHz, for 200 mm from 1 to 5.5 GHz, and for a block thickness of 300 mm, the subband is from

Advanced Construction Materials

33

1 to 3 GHz. The biggest difference between the parameters of S12 concrete and SF43 composite with a thickness of 100 mm is −52 dB at a frequency of 2.9 GHz, while the biggest difference between blocks with a thickness of 300 mm is −25 dB at the same frequency.

Fig. 10. Measurement results of the transmission parameters of concrete, concrete composite with steel (SF43) fibers in the frequency range 1 to 18 GHz of thickness: 100 mm; b) 200 mm and c) 300 mm.

In conclusion, it can be said that the concrete-based composite with steel fibers that shows significant reductions in EM wave energy transmission when passing through it.

4 Conclusion This paper deals with the electromagnetic wave (EM) transmission and reflection by concrete blocks focusing on the frequency range 30 MHz to 18 GHz. The results of the measurement of the reflection parameter show that there are no significant differences in the reflection coefficient S11 of the concrete block and the concrete composite with carbon fibers or with steel fibers. However, when comparing composites with carbon fibers and steel fibers via the transmission coefficients S12 , it is evident that composites with steel fibers provide additional attenuation compared to

34

V. Mandri´c et al.

composites with carbon fibers, on average about 15 dB for the range from 30 MHz to 1 GHz and about 18 dB for the range from 1 GHz to 18 GHz. The highest additional transmission attenuation of composites with steel fibers compared to pure concrete ranges from −33 dB (at frequencies 0.76 and 0.98 GHz) for a thickness of 100 mm to −22 dB (at frequencies 0.74 GHz) for a thickness of 300 mm. Furthermore, these largest additional reductions in the transmission parameter for the range above 1 GHz range from −52 dB (at a frequency of 2.9 GHz) for a thickness of 100 mm to −25 dB (at a frequency of 2.9 GHz) for a block thickness of 300 mm. These results indicate that there is a great potential for designing concrete composites with an increased ability to reduce the transmission of EM wave energy, with the ultimate goal of preserving the health and life of those who live in that space. Future research must include testing the transmission of EM wave energy when passing through concrete-based composites with different proportions of steel fibers while keeping the mechanical properties constant.

References 1. Halabe, U.B., Maser, K.: Propagation Characteristics of Electromagnetic Waves in Concrete, MIT Civil Engineering, US Army Research Office, March 1989 2. Stone, W.C.: Electromagnetic Signal Attenuation in Construction Materials, NISTIR 6055, NIST Construction Automation Program Report No. 3, Building and Fire Research Laboratory National Institute of Standards and Technology Gaithersburg, Maryland 20899, October 1997 3. Pinhasi, Y., Yahalom, A.: Propagation of ultra-wide-band signals in lossy dispersive media, Source: IEEE Xplore, June 2008. https://doi.org/10.1109/COMCAS.2008.4562803 4. NIST, EMF shielding by building materials, Attenuation of microwave band electromagnetic fields by common building materials, U.S. National Institute of Standards and Technology (NIST), 2012. (Updated 2017) 5. Mahfouz, M.R., Fathy, A., Badawi, A.: See-Through-Wall Imaging using Ultra Wideband Pulse Systems, Conference Paper, January 2005 6. Recommendation ITU-R P.2040-1 Effects of building materials and structures on radiowave propagation above about 100 MHz, P Series Radiowave propagation (2015). http://www.itu. int/ITU-R/go/patents/en 7. Kuruvilla, J., Runcy, V., Gejo, G.: Material for Potential EMF Shieldnig Application, 1 Edition. Elsevier, Amsterdam (2019) 8. Zhekov, S.S., Nazneen, Z., Franek, O., Pedersen, G.F.: Measurement of attenuation by building structures in cellular network bands. IEEE Antennas Wirel. Propag. Lett. 17(12) (2018)

Analysis of the Structure of Agricultural Machinery Repair Workshops - A Case Study Željko Baraˇc(B)

, Tomislav Juri´c , Ivan Plašˇcak , Monika Markovi´c , ´ c Antonija Koji´c , and Denis Cosi´

Faculty of Agrobiotechnical Sciences Osijek, Vladimira Preloga 1, 31000 Osijek, Croatia [email protected]

Abstract. Modern agricultural production implies the use of highly sophisticated machines. Their high reliability, efficiency, and operational life depend to a large extent on high-quality and timely regular preventive maintenance measures. Unfortunately, during exploitation, the machines will break down, and we attempt to restore them to their correct state with a high-quality, quick, and cheap repair. To achieve high-quality and efficient maintenance and repair, good work organization, necessary facilities, devices, and tools, as well as qualified and educated workers are essential. In the work, an audit of two central repair facilities for the repair of agricultural machinery was carried out to verify their advantages and disadvantages about the organization of maintenance and repair of machinery. The results of the research indicate the absence of diagnostics, maintenance planning, a longer time interval for the procurement of standard spare parts (3–5 days) in workshop 1, while in workshop 2 the only deficiency is manifested in the absence of maintenance planning. Keywords: Audit · Maintenance · Repair

1 Introduction A significant factor in agricultural production is the use of modern agricultural mechanization, primarily appearance tractors, and working machines made with a new concept of construction. To improve the exploitation of modern agricultural technology, it is crucial to evaluate factors that affect them during their usage [1]. Agricultural mechanization is indispensable in modern agricultural production because it enables the timely performance of agricultural operations, reduction of labor costs, efficient use of expensive raw materials (seed, fertilizer, protective means), improvement of product quality, improvement of soil productivity, and reduction the hard work of farmers [2]. Modern agriculture is characterized by a high level of mechanization, and one of the most important criteria of agricultural development is the level of agricultural mechanization in production [3, 4]. Historical development of maintenance and repair of agricultural machines is closely related to the development of machines production and their usage. During their exploitation, machines are exposed to various internal and external factors, which can cause © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 35–41, 2024. https://doi.org/10.1007/978-3-031-51494-4_4

36

Ž. Baraˇc et al.

their malfunction [5]. Agricultural mechanization is expected to have high exploitation reliability with maximum productivity but at the same time with minimal expenses of maintenance. Realization of these demands is possible with a responsible and prompt approach to service-predictive maintenance of agricultural machines. In fact, servicepredictive maintenance is an important factor in highly productive agriculture production [6, 7]. Maintenance of agricultural machines is crucial for their proper functioning, as well as for planning of production at family farms. Maintenance of agricultural machines aims to secure their functioning, attainability, and associated equipment in the given working conditions [8, 9]. During the agricultural season, lowered efficiency of the machines can result in the delay of completion of production as well as loss of crop yield and ineffective exploitation of work time. Malfunctions of tractors and other machines will result not only in great expenses for repair but also a decrease in the productivity of their work [10]. Predictive maintenance is a procedure that includes examination, service, and repair of machines according to the manufacturer. Predictive maintenance includes planned activities for machine maintenance and equipment, to extend the working lifetime of machines and to prevent unplanned repairs [11]. Corrective maintenance is a procedure that includes repair at the moment when malfunction occurs [12]. Irregular maintenance decreases tractor reliability, increases fuel consumption and emission of exhaust gases, and simultaneously decreases the power and service life of the engine [13]. Spare parts for the machines are essential for the quality of their maintenance and repair [14]. Machines effectivity increase considerably depends on their regular maintenance, which is the main task of the repair workshop. For the repair workshop to be effective essential is the structure of the workshop itself, with regard to choice of maintenance (predictive or corrective), equipment of the workshop with tools, management of spare parts, diagnostics, education of workers, etc. The term audit has a root in the Latin word “auditus” which means “I listen”. An audit is carried out to evaluate the work of individuals or organization. Audit of maintenance is carried out in order to evaluate the function of maintenance in the repair workshop. It is carried out by the technical director in the repair workshop with the help of the maintenance manager and external consultants [15]. There is a link between maintenance audit and measures of efficacy maintenance which are checked, in this case, these are benchmarking and backlog. Benchmarking is considered to be an efficient tool for the identification own performance compared to the competition. Backlog is considered to be future work (two weeks) on hold for every worker in the repair workshop of agricultural machinery and it is calculated based on the plan of maintenance. The low reliability of agricultural machines in operation is characterized by yield losses (due to the poor quality of work of certain parts) and high maintenance costs (frequent stoppages and repairs, increased consumption of fuel and lubricants, etc.). All this has a negative effect on the increase in the operating costs of machines, that is, on the increase in the unit prices of agricultural products [16, 17]. The research aims to determine the audit of the central repair workshop for the repair and maintenance of agricultural machinery, compare them, and provide guidelines for their improvement.

Analysis of the Structure of Agricultural Machinery Repair Workshops

37

2 Materials and Methods The research was conducted in two central repair workshops for the repair and maintenance of agricultural machinery. Furthermore, a maintenance audit was carried out to determine the state of Workshop 1 and Workshop 2, based on determined information, compare Workshop, and provide them guidelines for improvement of work and business. Measurement was conducted using a method of workshop condition determination and technical directors’ interviews. As stated before, a maintenance audit is conducted to evaluate function maintenance in the workshop using four measures: 1. 2. 3. 4.

Questionnaire preparation for audit, Interview of the technical director according to the prepared audit questionnaire, Interview results analysis and Report creation of audit.

During the workshops condition determination, necessary is to establish their condition realistically and objectively. The questionnaire for audit is displayed in tables. For the research, the survey method was used. To measure the results of the research, a Likert scale was used with a gradation of grades from 1 to 5. The answer to every question is evaluated on a scale of 1–5 grades and the average grades are obtained by summing the grades of maintenance in workshops. An important remark before questionnaire preparation for audit, necessary is to obtain general information on maintenance in workshops. For example: 1. 2. 3. 4. 5. 6. 7. 8. 9.

Which is an existent maintenance organization? How is preventive maintenance carried out? How is predictive maintenance carried out? How is maintenance programming carried out? How are the maintenance costs supervised? What is the system of work orders? How is the equipment classification carried out? How is the future work on hold (backlog) supervised? How is the management of spare parts carried out?

3 Results and Discussion Results are given in Table 1 and Table 2, and they are commented on and compared in the following text.

38

Ž. Baraˇc et al. Table 1. Maintenance audit conducted in Workshop 1

Question

Answer

Grades

Number of employees?

16 of these, 11 are mechanics, 1 for spare parts 4 and 4 in the administration with service and the director

Do you conduct staff educations?

Yes

How much vehicle fleet do you maintain?

150 tractors and 25 combine harvesters

5

Do you have an equipment register and how do you keep it?

We have, in the Jupiter program

5

Centralized, decentralized or combined maintenance?

Combined

5

5

What is maintenance strategy (concept)?

Maintenance according to service intervals

5

Do you have daily, weekly and long-term plans?

Annual plans

5

Are there preventive maintenance plans?

They exist, they are implemented

5

Are these plans being implemented?

No, only after reporting a malfunction

1

Are there plans for diagnostic examinations? We have, in the Jupiter program

5

Are these plans being implemented?

Yes

5

Do you have a work order system? How do you manage them?

Yes

5

Is a work order created for all jobs?

Yes

5

Are the rules for work order types being followed?

Yes

5

Are reports on the implementation of work orders submitted?

Based on the season of works, each type of machine before the season

5

Is the implementation of work orders being controlled?

The user of the machine reports a malfunction in the application (service portal). The malfunction is visible to all employees. The workshop manager assesses which mechanic goes out for intervention

5

How was maintenance programming performed?

No

1

What is the procedure from the occurrence of the malfunction to the repair?

Jupiter program

5

Are you planning an annual maintenance budget?

3–5 days

2

How are spare parts managed?

No

1

How long does the procurement process of standard spare parts take on average?

500 000 e

5

Do you make any spare parts yourself?

Yes

5

What is the total value of the stock of spare parts?

Yes

1

Is there an analysis of the effectiveness of spare parts management?

Yes

5

Do you hire external workers for some jobs? The problem of conscientiousness and education of tractor drivers, insufficient level

5

Total

105/125

Analysis of the Structure of Agricultural Machinery Repair Workshops

39

Table 2. Maintenance audit conducted in Workshop 2 Question

Answer

Grades

Number of employees?

25 of these, 13 mechanics, 4 salesmen, 2 for spare parts and 5 in the administration with service and the director

5

Do you conduct staff educations?

Yes

5

How much vehicle fleet do you maintain?

20 cars, 150 Claas tractors, 70 Claas combine harvesters and forage harvesters

5

Do you have an equipment register and how do you keep it?

We have, in the GATH program

5

Centralized, decentralized or combined maintenance?

centralized

5

What is maintenance strategy (concept)?

Planned maintenance, within prescribed intervals

5

Do you have daily, weekly and long-term plans?

Yes

5

Are there preventive maintenance plans?

Yes, they are implemented

5

Are these plans being implemented?

Yes, they are implemented

5

Are there plans for diagnostic examinations?

Yes, in the GATH program

5

Are these plans being implemented?

Yes

5

Do you have a work order system? How do you manage them?

Yes

5

Is a work order created for all jobs?

Yes

5

Are the rules for work order types being followed?

Yes

5

Are reports on the implementation of work orders submitted?

Depending on the amount of work, maintenance programs are made for the following days

5

Is the implementation of work orders being controlled?

- malfunction report

5

How was maintenance programming performed?

- malfunction diagnosis

1

What is the procedure from the occurrence of - selection of parts and orders the malfunction to the repair?

5

Are you planning an annual maintenance budget?

-installation and test

4

How are spare parts managed?

No

1

How long does the procurement process of standard spare parts take on average?

Using the warehouse Control program

5

Do you make any spare parts yourself?

1 day

5

What is the total value of the stock of spare parts?

No

1

Is there an analysis of the effectiveness of spare parts management?

265500–330500 e

5

Do you hire external workers for some jobs?

Yes

Total

5 105/125

40

Ž. Baraˇc et al.

Comparing the workshops, it was found that there was no diagnostics, no planning of the annual maintenance budget a longer time interval for the procurement of standard spare parts (3–5 days) in Workshop 1, while in Workshop 2 the only deficiency was the absence of planning the annual maintenance budget, which coincides with the research [14]. Furthermore, it is also visible that the total value of the stock of spare parts is higher at Workshop 1, where one person works on spare parts compared to Workshop 2, where the value of the stock of spare parts is lower and two workers are employed on spare parts, which is considered insufficient. According to the results of the research, Workshop 2 (116) received higher grades than Workshop 1 (105). Both workshops are satisfactory, and should increase the education of the workers themselves, plan the annual maintenance budget because with this they can plan future investments to improve the workshop’s operations. Workshop 1 needs to urgently introduce diagnostics because it is one of the basic structures within the workshop itself to increase the productivity and quality of the maintenance itself, which coincides with the research [5, 13, 18].

4 Conclusion The following conclusions were established by the conducted research: – Both workshops are satisfactory and achieved good grades (Workshop 1, 105 and Workshop 2, 116). – Compared to Workshop 2, Workshop 1 does not carry out diagnostics, does not plan the annual maintenance budget, and has a longer time interval for the procurement of standard spare parts (3–5 days), while Workshop 2 has only the deficiency in not planning the annual maintenance budget. – Workshop 1 needs to employ one more worker for spare parts due to the higher value of the stock of spare parts compared to Workshop 2 where two workers are employed for spare parts. – Both workshops should increase the number of education for the workers themselves to be more productive and have a higher quality of maintenance, and therefore more competitive on the market compared to other competitors. – Regarding the audit itself, both workshops should conduct it every year, and with that, they would see their deficiencies, i.e., adopt guidelines to be more competitive on the market in terms of productivity and quality of maintenance.

References 1. Brki´c, D., et al.: Exploitation of agricultural machinery. University of Josip Juraj Strossmayer in Osijek, Faculty of agriculture in Osijek, Osijek (2005) 2. Goyal, S.K., Prabha, P., Singh, S.R., Rai, J.P., Singh, S.N.: Agricultural mechanization for sustainable agricultural and rural development in Eastern U. P. - a review. Agric. Sustain. Dev. 2(1), 192–198 (2014) 3. Ozpinar, S., Cay, A.: The role of agricultural mechanization in farming system in a continental climate. J. Tekirdag Agric. Fac. 15(2), 58–72 (2018) 4. Banaj, Ð., Šmrˇckovi´c, P.: Management of agricultural machinery. University of Josip Juraj Strossmayer in Osijek, Faculty of agriculture in Osijek, Osijek (2003)

Analysis of the Structure of Agricultural Machinery Repair Workshops

41

5. Emert, R., Juri´c, T., Filipovi´c, D., Štefanek, E.: Maintenance of tractors and agricultural machinery. University of Josip Juraj Strossmayer in Osijek, Osijek (1995) 6. Juri´c, T., Emert, R., Šumanovac, L., Horvat, D.: The realization of machinery maintenance on family farms. In: Filipovi´c, D. (eds.) Proceeding, 29. International Symposium on Agricultural Engineering 2001, Opatija, Croatia, vol. 29, pp. 43–50 (2001) 7. Belak, S.: System of preventive or corrective maintenance, terotechnology. College for Tourism Management in Šibenik, Šibenik (2005) 8. Khodabakhshian, R.: A review of maintenance management of tractors and agricultural machinery: preventive maintenance systems. Agric. Eng. Int. CIGR E-J. 15(4), 147–157 (2013) 9. Plašˇcak, I., Juri´c, T., Emert, R.: Application of ferrography in condition based maintenance. Strojarstvo 52(2), 233–240 (2010) 10. Mishra, D., Satapathy, S.: Reliability and maintenance of agricultural machinery by MCDM approach. Int. J. Syst. Assur. Eng. Manag. 14, 135–146 (2021). https://doi.org/10.1007/s13 198-021-01256-y 11. Khodabakhshian, R., Shakeri, M.: Prediction of repair and maintenance costs of farm tractors by using of preventive maintenance. Int. J. Agric. Sci. 3(1), 39–44 (2011) 12. Majdandži´c, N.: Maintenance strategies and maintenance information systems. University of Josip Juraj Strossmayer in Osijek, Mechanical Engineering Faculty in Slavonski Brod, Slavonski Brod (1999) 13. Dahab, M.H., Gafar, M.A., Rahman, A.G.M.A.: Repair and maintenance cost estimation for two power sizes of agricultural tractors as affected by hours of use and age in years: a case study, Dongola area, Sudan. J. Eng. Res. Rep. 20(10), 113–121 (2021) 14. Segetlija, Z.: Distribution. University of Josip Juraj Strossmayer in Osijek, Faculty of Economics, Osijek (2006) 15. Plašˇcak, I., Baraˇc, Ž., Juri´c, T., Marinovi´c, D., Juratovi´c, I.: Audit central repair facility for repair of agricultural machinery – case study. In: Lackovi´c, Z. (eds.) Proceeding, 24th International Scientific Meeting Organisation and Technology of Maintenance 2015, OTO, Donji Miholjac, Croatia, vol. 24, pp. 157–162 (2015) 16. Desnica, E., Ašonja, A., Kljajin, M., Glavaš, H., Pastukhov, A.: Analysis of bearing assemblies refit in agricultural PTO shafts. Tech. Gazette 30(3), 872–881 (2023) 17. Ašonja, A., Desnica, E., Pastukhov, A., Kuznetsov, Y., Kravchenko, I.: Methode for quick determination of the reliability level of agricultural PTO shafts. In: Blaževi´c, D., Ademovi´c, N., Bari´c, T., Cumin, J., Desnica, E. (eds.) Proceeding, 31th International Scientific Meeting Organisation and Technology of Maintenance 2022, OTO, Osijek, Croatia, vol. 31, pp. 143– 149 (2022) 18. Lips, M., Burose, F.: Repair and maintenance costs for agricultural machines. Int. J. Agric. Manag. 1(3), 40–46 (2012)

Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o. ´ c(B) Denis Cosi´

, Željko Baraˇc , Tomislav Juri´c , Ivan Plašˇcak , and Josip Damjan

Faculty of Agrobiotechnical Sciences Osijek, Vladimira Preloga 1, 31000 Osijek, Croatia [email protected]

Abstract. The aim of this study is to examine the actual state of maintenance of agricultural machinery at Jerkovi´c d.o.o. The research results indicate that daily and weekly maintenance is carried out almost entirely according to the instructions for handling and maintenance. Technical protection and garage storage of the machines are per-formed adequately. However, to improve maintenance practices, machine operators need to be educated on the latest technologies. Keywords: Machinery · Maintenance · Repair workshop · Technical state · Failures

1 Introduction The world has an increasing need for food, leading to a greater implementation of agricultural machinery and a higher growth in agricultural production. Modern methods of land cultivation and agronomic practices require changes in agriculture, as well as newer and more powerful machinery to achieve higher efficiency. Agricultural production undergoes year-to-year changes, and there is a growing demand for reduced soil tillage. As a result, farmers are investing in newer, more sophisticated, and higher-quality machinery. Proper maintenance of agricultural equipment is one of the most significant factors in agriculture. A machine is most expensive when it is idle, and one of the main reasons for machinery failures is improper and untimely maintenance, along with poor storage conditions. The development of agricultural tractors is becoming increasingly complex day by day, so service technicians must undergo regular training to be able to maintain machines and expand their knowledge. Before starting work, the machine needs to be inspected according to the instructions for daily maintenance performed by the machine operator. Manufacturers provide a detailed description and instructions for daily, hourly, and weekly maintenance of the machine upon its handover. The authors [1] state that faster wear and increased maintenance costs in agricultural production are consequences of machines being exposed to negative atmospheric factors. Author [2] states that a good maintenance strategy is required for quality maintenance since it involves a larger number of employees. Authors [3–5] highlight the importance of conservation and protection of agricultural machinery and how corrosion can lead to © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 42–52, 2024. https://doi.org/10.1007/978-3-031-51494-4_5

Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o.

43

cracking and deterioration of certain components and assemblies, causing malfunctions. All machines during the winter dormant period should be stored in a closed, dry, and dark space. The damage caused to a tractor by being kept outdoors for one year is greater than the cost of constructing a simple shelter. Developed agricultural countries differ significantly from underdeveloped ones in terms of owners’ attitudes towards the storage and maintenance of machinery [6]. Authors [7] indicate that in developed and developing countries, proper maintenance plan selection has been the main requirement for improving the operational characteristics of agricultural machinery. The development of a maintenance plan and the selection of optimal machinery maintenance are facilitated by the application of multicriteria decision-making (MCDM). Excellent performance of all machine components and maximum efficiency can be achieved through preventive maintenance. The purpose of such maintenance is to continuously maintain all technical parameters at the desired level. By continuously monitoring service intervals and performing them, the service life of machines is extended, and larger failures are prevented [8]. Agricultural production is a specific activity, unlike the rest of the industry, where agricultural machines are often used to work in unfavorable working conditions and are engaged mainly seasonally [9]. The importance of maintenance can best be represented in internal combustion engines, where friction contributes significantly to mechanical losses, which make up 10–15% of energy [10]. The research aims to determine the maintenance of agricultural machinery in the company Jerkovi´c d.o.o. and provide guidelines for improvement.

2 Materials and Methods The research was conducted at Jerkovi´c d.o.o. where the technical condition of tractors, their equipment, and the quality of maintenance were assessed. Guidelines for improving machine maintenance were also provided. Jerkovi´c d.o.o. (Fig. 1) has been engaged in arable farming for many years. Initially, they were involved in livestock farming and cattle breeding, but in recent years, they have focused solely on arable production on 225 hectares of agricultural land. The history of Jerkovi´c d.o.o. dates back to 1958. When the Jerkovi´c family started introducing mechanization into agricultural production. With a series of successful years starting from 1970, they have been operating in the market as a sales and service partner and have been collaborating with Claas since 1978. After 2000, Jerkovi´c d.o.o. became a partner of Claas and grew into the main authorized partner for Eastern Croatia. They are also representatives of Horsch, Strautmann, Umega Agro, and the exclusive distributor of Trimble agricultural navigation systems for the entire Republic of Croatia. They own 6 tractors (Table 1) equipped with the Trimble guidance system. Daily and weekly maintenance is carried out by machine operators, while authorized company service technicians monitor and perform service intervals.

44

´ c et al. D. Cosi´

Fig. 1. Company Jerkovi´c d.o.o.

Table 1. List of agricultural machines. Type of machine

Year of manufacture

Machine power (kW)

Average time spent throughout the year (h)

Axion 960

2009

330

40

Axion 830

2014

168

300

Axion 830

2022

168

300

Arion 530

2021

107

60

Arion 430

2012

86

300

Case CS 105 Pro

2003

77

20

Lexion 6900

2020

373

100

Trion 650

2022

260

60

3 Results and Discussion In this chapter, the maintenance results of agricultural machinery in the researched company are presented and compared with the expert and scientific literature on maintenance.

Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o.

45

3.1 Claas Axion 960 From Table 2 it is evident that machine operators regularly perform daily maintenance of the machine according to the operating instructions, thereby extending its service life and reducing wear and tear [11]. The authors [8] state that daily maintenance should include checking brake functionality, engine oil levels, coolant level indicators, and visually inspecting the machine, which aligns with the findings of this research. Table 2. Daily maintenance of the Claas Axion 960 tractor. Name of daily maintenance activity

Conducted

Visual inspection of the machine

X

Cleaning the air filter

X

Checking the engine oil level

X

Checking the coolant level

X

Checking the measurementandcontrol instruments

X

Not conducted

From Table 3 it is evident that all the measures of weekly maintenance are performed on the investigated tractor, except for the inspection of wear and blockage of the lifting hitch and the checking of electrolyte level in the battery, as mentioned in [8, 11]. By carrying out all the regular measures, operators could directly contribute to maintenance cost savings [12–15]. Table 3. Weekly maintenance of the Claas Axion 960 tractor. Name of weekly maintenance activity

Conducted

Checking tire air pressure

X

Lubricating the tractor

X

Cleaning brake pads

X

Checking brake functionality

X

Washing and degreasing the machine

X

Not conducted

Checking wear and blockage of the lifting hitch

X

Checking the electrolyte level in the battery

X

46

´ c et al. D. Cosi´

3.2 Claas Axion 830 Company Jerkovi´c d.o.o. owns 2 Claas Axion 830 tractors and maintenance on both tractors is identical. All recommended actions regarding daily maintenance (Table 4) are carried out [16, 17]. Table 4. Daily maintenance of the Claas Axion 830 tractor. Name of daily maintenance activity

Conducted

Visual inspection of the machine

X

Cleaning the air cleaner

X

Checking the engine oil level

X

Checking the coolant level

X

Checking measurement and control instruments

X

Checking radiator cleanliness

X

Checking signaling

X

Checking hydraulic oil level

X

Not conducted

All weekly maintenance procedures (Table 5) of the tractor, except for checking wear and blockage of the lifting hitch and checking the electrolyte level in the battery, are performed regularly [16]. Table 5. Weekly maintenance of the Claas Axion 830 tractor. Name of weekly maintenance activity

Conducted

Checking tire air pressure

X

Lubricating the tractor

X

Cleaning brake pads

X

Checking brake functionality

X

Washing and degreasing the machine

X

Not conducted

Checking wear and blockage of the lifting hitch

X

Checking the electrolyte level in the battery

X

Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o.

47

3.3 Claas Arion 530 The daily maintenance of the Claas Arion 530 (Table 6) is carried out in its entirety [18]. Table 6. Daily maintenance of the Claas Arion 530 tractor. Name of daily maintenance activity

Conducted

Visual inspection of the machine

X

Cleaning the air cleaner

X

Checking the engine oil level

X

Checking the coolant level

X

Checking measurement and control instruments

X

Checking radiator cleanliness

X

Checking signaling

X

Checking hydraulic oil level

X

Not conducted

From Table 7 it is evident that all measures of weekly maintenance are performed, except for checking wear and blockage of the lifting hitch, as well as checking the electrolyte level in the battery, as prescribed by [8, 18]. Table 7. Weekly maintenance of the Claas Arion 530 tractor. Name of weekly maintenance activity

Conducted

Checking tire air pressure

X

Lubricating the tractor

X

Cleaning brake pads

X

Checking brake functionality

X

Washing and degreasing the machine

X

Not conducted

Checking wear and blockage of the lifting hitch

X

Checking the electrolyte level in the battery

X

3.4 Claas Arion 430 Claas Arion 430 is the tractor that is primarily used by the company Jerkovi´c d.o.o. throughout the year, and Table 8 presents the daily maintenance performed in its entirety, as prescribed by [19].

48

´ c et al. D. Cosi´ Table 8. Daily maintenance of the Claas Arion 430 tractor.

Name of daily maintenance activity

Conducted

Visual inspection of the machine

X

Cleaning the air cleaner

X

Checking the engine oil level

X

Checking the coolant level

X

Checking measurement and control instruments

X

Checking radiator cleanliness

X

Checking signaling

X

Checking hydraulic oil level

X

Not conducted

The weekly maintenance of the tractor, as shown in Table 9 is carried out completely, as indicated by [8, 19]. Table 9. Weekly maintenance of the Claas Arion 430 tractor. Name of weekly maintenance activity

Conducted

Checking tire air pressure

X

Lubricating the tractor

X

Cleaning brake pads

X

Checking brake functionality

X

Washing and degreasing the machine

X

Checking wear and blockage of the lifting hitch

X

Checking the electrolyte level in the battery

X

Not conducted

3.5 Case CS 105 Pro The weekly maintenance of the tractor is carried out according to the recommendations of the operation and maintenance manual [20]. Furthermore, since the tractor is not in operation every day, regular daily maintenance is not performed. However, the operators do not neglect the weekly maintenance of the tractor, as shown in Table 10.

Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o.

49

Table 10. Weekly maintenance of the Case CS 105 Pro tractor. Name of weekly maintenance activity

Conducted

Checking tire air pressure

X

Lubricating the tractor

X

Cleaning brake pads

X

Checking brake functionality

X

Washing and degreasing the machine

X

Checking wear and blockage of the lifting hook

X

Checking engine oil level

X

Cleaning the air filter

X

Cleaning the radiator

x

Checking coolant level

X

Checking battery electrolyte level

Not conducted

X

3.6 Claas Lexion 6900 and Claas Trion 650 The combine harvester undergoes thorough maintenance before the harvesting season. The combine harvester is a specific machine, and its maintenance is carried out before and after the harvest season. As shown in Table 11 maintenance is performed regularly and according to the operation and maintenance manual instructions. Repair and maintenance of the combine harvester are conducted from December 1st to April 15th when the machine is not in use. The combine harvester is used for approximately 90 days per year and operates in specific and varied working conditions. Therefore, great attention must be paid to its repair and maintenance. The most common malfunctions of the combine harvester are a result of overloading and inadequate maintenance [8]. Table 11 and Table 12 demonstrate that the Claas Lexion 6900 and Claas Trion 650 combines are regularly maintained according to the recommendations from the operation and maintenance manual, as well as scientific and professional literature [21, 22].

50

´ c et al. D. Cosi´ Table 11. Maintenance of the Claas Lexion 6900 harvester.

Name of maintenance activity

Conducted

Checking fuel system for leaks

X

Checking diesel engine oil level

X

Checking diesel engine coolant level

X

Inspecting coolant hoses

X

Inspecting air hoses

X

Checking brake fluid level

X

Checking engine oil levels

X

Checking hydraulic system for leaks

X

Performing pre-harvest calibration procedures

x

Checking oil level of the radial spreader drive

X

Checking oil level of the grain tank discharge drive

X

Not conducted

Table 12. Maintenance of the Claas Trion 650 harvester. Name of maintenance activity

Conducted

Fuel system leak check

X

Diesel engine oil level check

X

Diesel engine coolant level check

X

Coolant hose inspection

X

Air hose inspection

X

Brake fluid level check

X

Engine oil level check

X

Hydraulic system leak check

X

Pre-harvest calibration procedures

x

Radial spreader drive oil level check

X

Grain tank discharge drive oil level check

x

Not conducted

4 Conclusion Based on the conducted research, the following findings have been determined, • Tractors and combines are equipped with all the necessary documentation. • The farm invests in the renewal of their machinery each year, acquiring the latest tractors and technologies.

Maintenance of Agricultural Machinery in the Company Jerkovi´c d.o.o.

51

• Regular daily and weekly maintenance procedures for tractors and combines are being carried out, with the exception of the weekly maintenance tasks related to checking the wear and blockage of the lifting hitch and inspecting the electrolyte level in the battery. • Machine operators are educated and trained in performing daily and weekly maintenance tasks. It is important to emphasize that a thorough inspection of the machines during daily and weekly maintenance is crucial, and each procedure is of utmost importance. The employees of the company have been made aware that they should not neglect the task of checking the electrolyte level in the battery, as it is often overlooked, as well as checking the wear and blockage of the lifting hitch. The manufacturer of each tractor and combine sets forth the measures for daily and weekly maintenance for a reason, and therefore, they must be implemented to prolong the machine’s lifespan.

References 1. Banaj, Ð., Šmr´ckovi´c, P.: Management of agricultural machinery, Osijek, 162, 24 (2003) 2. Bekˇci´c, M.: Maintenance and Overhaul of Machinery. Textbook, Belgrade (1981) 3. Baraˇc, Ž., Juri´c, T., Plašˇcak, I., Heffer, G., Kramer, M.: Organization and importance of service preventive maintenance in “PP Orahovica” with regard to environmental protection. In: Proceedings of the 25th Scientific-Professional Conference, OTO 2015, Osijek, pp. 51–56 (2016) 4. Juri´c, T., Emert, R., Šumanovac, L.: Implementation of maintenance measures on family farms. In: Proceedings of the “Current Tasks of Agricultural Mechanization” Conference, Opatija, pp. 43–49 (2001) 5. Landeka, S.: Engines and Tractors. Textbook, Vinkovci (1995) 6. Economic Journal. https://gospodarski.hr/rubrike/mehanizacija/spremanje-strojeva-prekozime/2023/04/04 7. Mishra, D., Satapathy, S.: Reliability and maintenance of agricultural machinery by MCDM approach. Int. J. Syst. Assur. Eng. Manag. 14, 135–146 (2021). https://doi.org/10.1007/s13 198-021-01256-y 8. Emert, R., Juri´c, T., Filipovi´c, D., Štefanek, E.: Maintenance of tractors and agricultural machinery, 126 p. Josip Juraj Strossmayer University, Faculty of Agriculture, Osijek (1995) 9. Ašonja, A., Desnica, E., Pastukhov, A., Kuznetsov, Y., Kravchenko, I.: Methode for quick determination of the reliability level of agricultural PTO shafts. In: Blaževi´c, D., Ademovi´c, N., Bari´c, T., Cumin, J., Desnica, E. (eds.) OTO 2022, vol. 592, pp. 143–149. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-21429-5_13 10. Janji´c, N., Adamovi´c, Ž, Nikoli´c, D., Ašonja, A., Stojanovi´c, B.: Impact of diagnostics state model to the reliability of motor vehicles. J. Balkan Tribological Assoc. 21(2), 511–522 (2015) 11. Operator’s Manual for Operation and Maintenance of Claas Axion 960 (2009) 12. Sopegno, A., Calvo, A., Berruto, R., Busato, P., Bocthis, D.: A web mobile application for agriculture machinery cost analysis. Comput. Electron. Agric. 130, 158–168 (2016) 13. Bochtis, D.D., Sørensen, C.G.C., Busato, P.: Advances in agricultural machinery management: a review. Biosyst. Eng. 126, 69–81 (2014) 14. Najafi, B., Torabi Dastgerduei, S.: Optimization of machinery use on farms with emphasis on timeliness costs. J. Agric. Sci. Technol. 17(3), 533–541 (2015)

52

´ c et al. D. Cosi´

15. USDA. Characteristics and production costs of U.S. corn farms, including organic. Economic Research Service, United States Department of Agriculture, Washington, D.C. (2010) 16. Operator’s Manual for Operation and Maintenance of Claas Axion 830 (2014) 17. Operator’s Manual for Operation and Maintenance of Claas Axion 830 (2022) 18. Operator’s Manual for Operation and Maintenance of Claas Arion 530 (2021) 19. Operator’s Manual for Operation and Maintenance of Claas Arion 430 (2012) 20. Operator’s Manual for Operation and Maintenance of Case CS 105 Pro (2003) 21. Operator’s Manual for Operation and Maintenance of Claas Lexion 6900 (2020) 22. Operator’s Manual for Operation and Maintenance of Claas Trion 650 (2022)

Bridging the Physical and Virtual Worlds: A Hand Tracking Gesture Recognition System for XR Applications Matija Fumić and Časlav Livada(B) Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Josip Juraj Strossmayer University in Osijek, Kneza Trpimira 2B, 31000 Osijek, Croatia [email protected]

Abstract. Extended reality (XR) is a term that encompasses augmented reality (AR), virtual reality (VR), and mixed reality (MR). These technologies allow users to interact with digital content in immersive and realistic ways. One of the challenges of XR is to provide natural and intuitive ways to control and manipulate the virtual environment. This paper presents the development of a gesture recognition system for hand tracking in extended reality technologies (XR) and its practical applications. The system uses a gesture definition language that allows the user to create custom gestures based on the hand skeleton model. The system was implemented using the Oculus Quest 2 device and the Oculus SDK in the Unity game engine and is compatible with all devices that support hand tracking. The gesture recognition system was implemented using a three-bone thumb and little finger model that is compatible with all hand models that contain four bones. The implementation of this gesture recognition system in XR has several benefits, including increased immersion, improved interaction with virtual objects, and reduced physical fatigue compared to traditional input methods such as controllers or keyboards. This system provides users with a more intuitive and natural way to interact with their virtual environment. Overall, this work demonstrates the feasibility and potential of integrating gesture recognition into XR technologies and its ability to enhance user experience and interaction. The system presented in this work can serve as a foundation for future developments in this area, with potential applications in gaming, education, healthcare, and other fields. Keywords: Augmented reality · Extended reality recognition · Hand tracking · Mixed reality

1

· Gesture

Introduction

The acronym XR opens the door to a wide range of different technologies and possibilities. Extended Reality includes a set of three different but similar technologies. The first of these, and also the most widely used, is referred to by the c The Author(s), under exclusive license to Springer Nature Switzerland AG 2024  T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 53–74, 2024. https://doi.org/10.1007/978-3-031-51494-4_6

54

M. Fumić and Č. Livada

acronym AR (Augmented Realilty) and represents an extension of reality by another layer. This means that another digital media layer is added on top of the reality layer that the user sees with the camera. Most often this is models, images, videos and other multimedia content. This type of content is mostly presented through mobile and web applications. Although AR glasses exist, they are not yet very widespread, partly because of the price and partly because almost everyone already has their own AR device in the form of their cell phone. It is most commonly used for educational purposes, as it makes it easier for students to visualize some inaccessible things. An example of AR technology can be seen in Fig. 1.

Fig. 1. Display of AR technology.

The next addressed technology is known as Mixed Reality and is the least common type of augmented reality, where the two levels mentioned above (reality

Bridging the Physical and Virtual Worlds

55

and digital) merge. These two technologies are often confused and difficult to distinguish from each other. The main difference is that AR is a layer of the digital over the real and such digital objects do not have the ability to interact with the real world. In MR, these two layers are mixed into one, which opens the possibility of interaction of digital objects with the real world. This technology is not as widely used because it is not available and is very expensive. Currently, the only real MR devices are the Microsoft Hololens and the Microsoft Hololens 2 [1]. The easiest way for people to understand this technology is to draw a parallel with holograms. Illustration of holograms in mixed reality is shown on Fig. 2.

Fig. 2. Screenshot from Microsoft Hololens.

The last of the three technologies belonging to this group, and at the same time the one chosen for the presentation of this work, is virtual reality, i.e. the technology of full immersion in the digital world. Unlike the previous two technologies, in virtual there is no level of the real world, but everything takes place in a completely digital world. The user is fully immersed in the virtual world and has full power to control virtual objects. An example of virtual reality technology can be seen in Fig. 3. This technology finds its greatest application in the video game industry, but is also present in tourism, education, and virtual training. It finds its beginnings

56

M. Fumić and Č. Livada

Fig. 3. Screenshot from Oculus Quest.

mainly in so-called virtual walks, where users can experience an area in a different guise, as it looked two hundred years ago, or visit a desired area such as the Pyramids of Giza. As the technology and virtual reality devices themselves have evolved, so have the capabilities of such applications. To better understand the timeline of this technology, this example uses Oculus Android devices. The first of these devices was Oculus Go [2]. A device with three degrees of freedom (rotation around the x, y, and z axes). It comes with a single controller and is not powerful enough to display more complex applications. It was used precisely for the virtual walks and 360-degree video display mentioned above. After it, the turning point in the world of VR came with the Oculus Quest, i.e. its successor, the Oculus Quest 2, currently probably the most popular Android VR device. It has six degrees of freedom. In addition to the three mentioned above, the Quest has three more spatial coordinates, which means that the user can move around in the real world. This device is powerful enough to run demanding content, which is why it is widely used in the gaming industry. Another important application enabled by these devices is training employees for high-risk or life-threatening situations. It is difficult to immediately introduce employees to the operation of sophisticated machines that can cost them an injury or their lives, so they should first be prepared by an application that credibly shows what to expect when they operate mentioned machine. Of course, there are also devices that are

Bridging the Physical and Virtual Worlds

57

only used to display the image, while all the logic is processed on the computer. Such devices do not have their own operating system and must be physically connected to a computer. The devices in themselves immerse the user in the digital world, but there are still attempts to raise this feeling to a higher level. This is where handheld control comes into play. Until recently, the only way to manage content in the virtual world was to use a controller. The integration of infrared cameras into devices that recognize human hands has opened up a new world of possibilities. Hand interaction is much more intuitive and natural. Users find their way around applications more easily, and they do not have to go through training on managing applications or anything like that. In everyday communication, people are used to unconsciously (or consciously) gesturing with their hands. And so gestures, such as those of the open palm, fist, thumb down, or thumb up, find their place in daily life. The task of this work is to implement a gesture recognition system in such applications and allow users to easily associate their functionalities with a specific gesture and make their applications more intuitive (for example: Highlighting places where the user points his finger, etc.).

2

Overview of the Field of Gesture Recognition in Augmented Reality Technologies

Hand recognition using computer vision has been around for a number of years. There are a large number of libraries that perform this task, but currently the most widely used is MediaPipe from Google [3]. This is an open source machine learning library for live and streaming media (both real-time and pre-rendered images). It is most commonly used with the OpenCV environment and the Tensorflow machine learning library. MediaPipe comes with a pre-learned hand detection algorithm [4]. OpenCV is used for real-time image recognition, while Tensorflow is used for hand gesture recognition (thumbs up, thumbs down...). A representation of the hand with the MediaPipe library can be seen in Fig. 4. Handtrack.js is a javascript library for real-time hand tracking in a web browser. It contains a pre-learned model using Tensorflow and a convolutional neural network. The selected dataset is Egohands from Indiana University. This library recognises basic gestures such as open palm, closed palm, and pinch [6]. ManoMotion MobileAR is a software development kit (SDK) for developing Unity AR applications with the ARFoundation upgrade. It provides hand recognition and basic gestures using the camera on Android and iOS devices. This is a new project and there is not much information about it yet, but it is worth mentioning as a potential technology in this area [7,8]. The above libraries are used in a variety of technologies, but none of them are appropriate for virtual reality technology. Oculus (now Meta) didn’t introduce the use of hands to control the device until December 2019. At that time, hand tracking was only available in their menu, and it wasn’t until late 2020 that the first apps supporting hand tracking were released. Today, the hand tracking approach is available to all in Oculus’ SDK. The first concept of gesture recognition

58

M. Fumić and Č. Livada

Fig. 4. Arms and arm bones using the MediaPipe library [5].

in this technology was presented by professor Quentin Valembois from the University of Liège, Belgium [9]. He took the basic idea for this work and developed it into the entire gesture recognition system. Two years after Quentin’s work was published and during the development of this work, Oculus added a section called Pose Detection to its SDK. The SDK allows the user to define the state of each finger (extended, partially bent, fully bent) and create a gesture with such a combination, so that when the thumb is raised, the thumb is extended and all other fingers are fully bent [10].

3

Technological Tools and Platforms for the Experiment

This chapter will describe the technologies and tools used in the preparation of this paper. 3.1

Unity Game Engine

Unity is a cross-platform real-time 3D/2D game engine. Cross-platform means that applications created in Unity can be exported to different platforms (Windows, Android, iOS, PlayStation, Server, Web...). This is possible because the main scripting language in Unity is C# and the specific implementation for Unity is based on Mon and IL2CPP [11]. Realtime means that the app is displayed in real time. Movies (generally videos) are created, generated and their display is never in real time. Unity applications are displayed in real time, which opens the possibility of various interactions and reactions to them. Unity provides the ability to create applications in three of these two dimensions. The work will be implemented on the LTS version of Unity 2020. IL2CPP backend, NET Standard

Bridging the Physical and Virtual Worlds

59

2.0 and Roslyn translators have been used. The project is adapted to export the app to Android, since Oculus Quest is based on Android. Unity’s interface makes it easier for developers to develop in an intuitive way and the game engine itself is designed to allow the developer to focus more on the game mechanics themselves than on implementing systems for physics, collisions, and similar. Unity’s graphical interface is shown in Fig. 5.

Fig. 5. Unity Graphical User Interface.

3.2

Oculus Quest 2

Oculus Quest 2 is the third in a series of Oculus Android devices, the second in the Quest series. It was launched in late 2020. The biggest improvement over its predecessor is Qualcomm’s revolutionary Snapdragon XR2 processor, designed exclusively for virtual reality devices. In terms of processing power, it is most similar to the Samsung Galaxy S20, but with one major difference. Quest 2 (Fig. 6) has to display an image on two 1832 × 1920 screens, one for each eye, which requires a stronger GPU. It comes with two controllers that are connected to the device via Bluetooth. As mentioned earlier, the controllers can also be replaced with actual hand interactions that the device detects using four infrared cameras on the front of the device. This device is affordable and has a strong presence in the VR device market, and its advanced hand interaction capabilities makes it an ideal candidate for this papers demonstration [12]. As mentioned earlier, Oculus Quest has four infrared depth cameras to monitor the state of the environment. With the help of these four cameras and inertial measurement units that use VISLAM (Visual - Inertial Simultaneous Localization and Mapping) technology, the device determines its exact position in space and its position relative to other objects. In this way, the device creates point

60

M. Fumić and Č. Livada

Fig. 6. Oculus Quest 2.

clouds of its surroundings at a real-time frequency of 30 frames per second. Using these point clouds and machine learning algorithms, the user’s hands are successfully detected. Oculus creates a skeleton of the arm, as shown in Fig. 4. Using the API (Application Programming Interface) in Unity, you can access the local rotations of each bone (note that each finger is a nested bone structure and the distal phalanx is the child of the medial phalanx in the hierarchy, which in turn is the child of the proximal phalanx in the hierarchy, and so on.) and plot them in your skeleton and fitted mesh hands (3D dataset consisting of polygons contained in x, y, and z reference points describing the height, width, and depth of the model). For the purposes of this work, an input data adapter will be written to provide a standardized data output. This means that the source code of this work can be used on other devices such as the HTC Vive Focus or Leap Motion. It is only necessary to give the adapter an exact set of input data. 3.3

Oculus SDK

To make it easier for developers to create apps for their devices, Oculus has developed its own SDK. Since it is possible to create apps for multiple platforms, there are four types of SDKs for four platforms. Specifically, there is an SDK for web platforms, the original Android development, the Unreal Game Engine, and the Unity Game Engine. This paper experiment was conducted in Unity, so the Oculus integration package was taken from the Unity store. At the time of writing this paper, version 42 is current, and this was used for the build. 3.4

OctoXR

OctoXR is a tool that can be purchased from the Unity Store. It comes with prebuilt tools for creating a VR application [13]. This work was created as part of OctoXR and all source code is available in the store. Besides gesture recognition, this tool also includes basic VR interactions such as grasping things, moving, avatar animations, user interface interaction and similar. Its goal is to make it

Bridging the Physical and Virtual Worlds

61

easier for users to create such applications. Developers should no longer focus on implementing basic functions, but on the quality of the application itself and the mechanics.

4

Gesture Detection Algorithm

The gesture recognition system consists of several key components such as creating a gesture, storing a gesture, importing concrete gestures into the scene, and recognizing the gesture itself. In this chapter, the implementation and application of the algorithm is described step by step. Creating a gesture is performed through the GestureCreator.cs and it is important that it is on the same object as the root element of the hand skeleton so that it has access to the entire skeleton. The root element contains a list of all the bones in the hand. When the method is called, it iterates through the list of all bones and calculates the relative position of each bone with respect to the root bone and stores it in a new sheet. This can be represented as placing the root bone in a coordinate (0, 0, 0) and calculating what local position each bone is in with respect to the origin of the new coordinate system. Figure 7 shows the relative position of the distal phalanx of the middle finger of the left hand and the bone roots (wrist).

Fig. 7. The relative position of the distal phalanx of the middle finger.

62

M. Fumić and Č. Livada

The Unity game engine uses a left-sided coordinate system with an upward yaxis. Each object in Unity also has its own local coordinate system. If this object is on a rotation of 0 around the x-axis, 0 around the y-axis, and 0 around the zaxis, then its local coordinate system matches the global one. When calculating the local positions of the bones relative to the root bone, the scalar product of each coordinate axis of the root bone’s local coordinate system with the unit vector of the vertical vector up (0, 1, 0) is calculated. This is performed to avoid false detection of gestures. If gestures were detected based only on the local positions of the bones with respect to the root bone, thumbs up and thumbs down would be one and the same gesture. Therefore, the orientation of the hand itself, i.e., the root bone, must also be taken into account. The vertical unit vector upwards is taken as the reference vector because it is always the same. The unit vector to the right may or may not be the same when creating a gesture; more precisely, their scalar product does not have to be the same. The scalar product of two vectors can be represented as a measure of the correspondence (value from 0 to 1) of two vectors. More specifically, the projection of one vector onto another. The scalar product is shown in Fig. 8, and Eq. 1 shows how the scalar product is calculated. a · b = |a||b|cosθ

(1)

Fig. 8. Scalar product of two vectors [14].

The x-axis of the local coordinate system of the root bone (wrist) should be in the direction of the thumb (red axis in Fig. 9). For a thumb-up gesture, the scalar product of the local x-axis and the global unit vector upward gives a value of approximately 1 (they have approximately the same direction and orientation). For the thumbs-down gesture, the scalar product of the local xaxis and the global unit vector up gives a value of approximately −1 (they have approximately the same direction but opposite orientations). For this reason, the scalar product of all three local coordinate axes is monitored and the global orientation of the hand in space can be determined from these three data.

Bridging the Physical and Virtual Worlds

63

Fig. 9. Local coordinate axes of the root bone of the hand.

After calculating the local bone positions and scalar products, a script object is created in which these values are recorded along with other data such as the name of the gesture, hand, and similar. Such a script object can then be saved and used in other scenes/projects/applications. This is possible because the GestureManager.cs component can import several different gestures via the menu. This component is responsible for unifying all gestures for a scene and linking them to their functionalities. Script objects are translated into a structure that can be serialized (the process of converting structures into a format that Unity can store and later reconstruct when executed) and displayed in the editor, and each gesture has three events associated with it: Gesture detected, Gesture detected but with wrong rotation, Gesture lost. How this looks in the Unity menu can be seen in Fig. 10.

64

M. Fumić and Č. Livada

Fig. 10. Display of linking functionality with specific gestures.

The GestureDetector.cs component is also closely related to the root element of the hand skeleton, as it requires access to all bones. When it is started, it retrieves all gestures and events related to that hand and compares the current state of the hand with the created gestures in each frame of the application. This can be done in two ways. The first way is continuous detection. Each frame looks for a new gesture, and when it is found, it calls certain functions subscribed to that event. That is, when the user gives the thumbs-up gesture, the function related to the thumbs-up gesture is called in each frame. Another option is discrete recognition. In this form of recognition, gestures are recognized in each frame, but the functions related to a particular gesture are called only once, at the moment it is first recognized. That is, when the user points the thumb up, the functions related to that gesture are called only when it is first recognized, and are not called again. The following functions are called only when another gesture, different from the previous one, is detected. Each image is passed through the list of created gestures. For each gesture, all bones are run through again. During this process, the current local position of the bone with respect to the root bone is calculated. If the current local position of the bone is approximately the same as the one that was saved for this bone when this gesture was created, the difference in positions (variable x) is saved and the next bone is moved. If the difference

Bridging the Physical and Virtual Worlds

65

in bone positions is too large, this gesture is discarded and the next gesture is executed. When the check of all bones for a gesture is completed, it means that all bones are approximately in the positions required for that gesture. Any differences in the positions of the bones are added up (sum of all variables x for each bone) and stored (variable y). When going through all the created gestures, the gesture that has the smallest sum of position differences (the smallest variable y) is recognized. If no gesture is detected, the algorithm returns an empty gesture. This one gesture (detected or empty) is then used for further calculations. First, it checks if the returned gesture is equal to an empty gesture. If an empty gesture is detected and a specific gesture was detected in the previous frame of the application, this is an indication that all functions associated with the gesture event for the previous gesture are lost, and the iteration of the algorithm ends in that frame. However, if a gesture other than an empty gesture is detected, the iteration continues. Then it is checked if the discrete mode is selected and the current gesture is equal to the gesture detected in the last frame, and the iteration for that frame is stopped. If the mode is continuous or discrete, but the detected gesture is different from the previous one, the algorithm continues. The instantaneous scalar product of the local coordinate axes of the root bone with the global unit vector up is calculated. If these three scalar products approximate those stored for that gesture, the detected gesture is considered to have a good orientation. If the scalar products do not approximately match those stored, it is assumed that the gesture was detected, but with an incorrect rotation. According to this decision, certain functionalities are called, which are assigned to one of the two events. It is not possible to recognize both events.

5

Algorithm Implementation and Application

In this chapter, two examples of algorithm implementation are described. The first example is Rock, Scissors, Paper, a popular game in which two players simultaneously choose between rock, scissors, or paper. The second example is Gesture Hero, a game that uses an algorithm to create an interactive experience. Players must mimic certain gestures that are detected and analyzed. 5.1

Rock, Scissors, Paper

Rock, Scissors, Paper is an iconic game that revolves around three different gestures, each representing a different object. The rock gesture, the hand is clenched into a tight fist, moving all three joints of all five fingers. In the scissors gesture, on the other hand, the joints of the little finger, ring finger, and thumb are flexed, and the joints of the middle finger and index finger are extended. Finally, in the paper gesture, all five fingers of the hand are fully extended.

66

M. Fumić and Č. Livada

In this engaging application, users compete against the computer in an exciting battle. The computer generates a random gesture from the three available options, and the user must quickly repeat their own gesture after the time limit expires. The game centers on the strategic clash of gestures, with each gesture being superior to another: rock triumphs over scissors, scissors cuts through paper, and paper envelops rock. To determine the final winner, a best-of-five series is played. The first player to win three games is the overall winner. Figures 11 through 15 visually depict the various phases and outcomes of the game, illustrating the excitement and intensity of the game experience. Figure 11 shows the menu screen that allows players to start the game. The menu has intuitive buttons or prompts so that users can easily navigate and select their desired game mode. In Fig. 12, we see a drawn game state where both the player and the computer have chosen the rock gesture. This image shows the moment of tension just before the outcome is determined. The player’s hand clenched into a fist reflects the gesture of the rock, while the gesture chosen by the computer remains unknown. This creates a sense of anticipation as players anxiously wait for the computer’s move to be revealed. Figure 13 captures a moment of defeat for the player, as he chose the rock gesture while the computer chose paper. The image shows the clash of the two gestures and symbolizes the player’s defeat in this round. This result shows the superiority of the paper over the stone, since it embraces and defeats the stone’s gesture. In contrast, Fig. 14 represents a victorious moment for the player. By revealing the paper gesture while the computer displays rock, the player emerges victorious in the round. The image visually represents the triumph of the player’s strategy as the paper gesture triumphs over the rock gesture. Finally, Fig. 15 shows another triumph of the player. This time the player chooses the scissors gesture while the computer chooses the paper gesture. The image illustrates the success of the player’s strategic decision: scissors triumphs over the paper gesture. This result illustrates the player’s ability to outsmart the computer and achieve victory.

Bridging the Physical and Virtual Worlds

Fig. 11. Display of the menu to start the game.

Fig. 12. Undecided game. The player and the computer showed the rock.

67

68

M. Fumić and Č. Livada

Fig. 13. The player lost by showing the rock and the computer showed the paper.

Fig. 14. The player won by showing the paper and the computer showed the rock.

Bridging the Physical and Virtual Worlds

69

Fig. 15. The player won by showing the scissors and the computer showed the paper.

5.2

Gesture Hero

Gesture Hero is an exciting application inspired by the extremely popular game Guitar Hero. Designed to captivate players, this interactive game puts users in a fascinating space environment. In this virtual world, two lines of circles gracefully approach the player, with one line aimed at the left hand and the other at the right. Each circle features a special gesture that makes for a visually appealing experience. To succeed in Gesture Hero, players must align their hands with the incoming circles and accurately mimic the gestures depicted. As the game progresses, the pace picks up: the circles race towards the player at an ever-increasing speed. In addition, the number of circles multiplies, requiring maximum concentration, skill and coordination from the player. At the end of the game, the player’s performance is evaluated based on his accuracy in executing the gestures. This evaluation culminates in the awarding of a medal symbolizing the player’s skill and achievement. The type of medal received is determined by the percentage of successfully executed gestures during the game. Figure 16 through Fig. 20 vividly illustrate the game mechanics and provide a visual representation of Gesture Hero’s immersive gameplay experience. Figure 16 illustrates the first step of starting the game in Gesture Hero. It shows the process of detecting two overt palm gestures from the user. The image shows a user interface where the user’s hands are positioned in front of a camera and can be detected. This step is used to set up the game and ensure that the

70

M. Fumić and Č. Livada

user’s hand gestures can be accurately detected and registered in the game. In the Fig. 17 we can see how the gestures approach the user. The image shows the circular icons that represent gestures moving directly toward the user. These icons can appear on a virtual display, creating an immersive experience where the gestures seem to emerge from the depths of the screen towards the user’s perspective. This visual representation adds to the immersive nature of the game. Figure 18 captures a moment of success for the user. The image shows that the user has correctly mimicked both gestures by placing their hands in the corresponding circles and demonstrating the gestures depicted on them. This successful execution indicates that the user has correctly and skillfully mimicked the gestures, which earns them points. In Fig. 19, the image shows upcoming gestures for which the user must prepare. It likely shows a preview of the gestures that will appear in subsequent rounds, giving the user a brief moment to prepare and strategize for the challenges ahead. This image gives a glimpse of the fast-paced nature of Gesture Hero, where the user must quickly adapt to changing gestures and maintain their accuracy. Finally, Fig. 20 represents the end of a game in Gesture Hero. The image symbolizes the end of the game and displays a summary of the user’s performance, score, or achievements. It indicates that the user has reached the end of their game session and provides closure to the game experience. Overall, these images provide insights into the game mechanics and gameplay of Gesture Hero. They show gesture recognition, successful execution by the user, anticipation of upcoming challenges, and completion of a game session.

Fig. 16. Start the game by detecting two open palm gestures.

Bridging the Physical and Virtual Worlds

Fig. 17. Gestures come directly in front of the user.

Fig. 18. The user correctly showed both gestures.

71

72

M. Fumić and Č. Livada

Fig. 19. Show upcoming gestures.

Fig. 20. The end of a game.

Bridging the Physical and Virtual Worlds

6

73

Conclusion

Augmented reality technologies are still relatively young and have not yet reached their full potential. This is mainly because such devices are expensive and difficult to reach for the general public. A revolution in the market for such devices was sparked by Meta (then Facebook) with its Oculus Quest 2 device. Its price is currently the same as that of the PlayStation 5, and the number of supported applications is growing. Since it is an Android device, it supports augmented reality apps as well as most regular Android apps. In July 2021, Oculus introduced a passthrough that upgraded this device from supporting only VR content to MR and AR content. As more devices become available, it is important that they are as comfortable and easy to use as possible for users. Using your hands is a big advantage here. Almost each one of these devices comes with its own controllers. These can be used in the home menu, applications, games and similar. What they all have in common is that when each application is launched, it must be explained in some way how to use that application. If you use your hands, this part can be practically omitted. Hands are intuitive. They make interactions easier and more natural. To grab something, you do not have to press a button, just make a grabbing gesture. Users who wear glasses on their heads for the first time have a hard time navigating the virtual world. Natural interactions with their hands are exactly what helps them adapt to and interact with virtual environments more quickly and easily. People frequently gesture with their hands in everyday life. Whether unconsciously waving, warning of danger, nonverbal communication, or sign language, this type of communication has become increasingly common. Recognizing these gestures is another layer to bring the virtual world closer to people and make it even easier and more user-friendly. By using this gesture recognition system, the user can very easily associate some functions with a certain gesture and speed up the development of the application. This system can be further extended. All the work is based on static hand gestures that are independent of each other. It remains to implement a system for dynamic recognition of gestures, such as waving, indicating that someone is approaching, and similar. Such a gesture is a combination of a static gesture and the speed of movement of the hand itself in different directions. Then it remains only to realize a combination of two dynamic gestures with two different hands. The best example of such a gesture would be showing traveling in basketball or handball. It is a gesture in which the fingers of both palms are fully flexed and the hands perform a twisting motion around the other hand. In addition to improving the gesture system, the efficiency of the algorithm must also be improved and ways must be found to avoid difficult mathematical operations (since this algorithm is executed at every frame of the application).

74

M. Fumić and Č. Livada

References 1. Hand tracking gesture detection. https://www.microsoft.com/en-us/hololens. Accessed 8 Nov 2022 2. Metaquest oculus go. https://www.oculus.com/experiences/go/. Accessed 12 Nov 2022 3. Lugaresi, C., et al.: MediaPipe: a framework for building perception pipelines. arXiv preprint arXiv:1906.08172 (2019) 4. Zhang, F., et al.: MediaPipe hands: on-device real-time hand tracking. arXiv preprint arXiv:2006.10214 (2020) 5. On-device, real-time hand tracking with mediapipe. https://ai.googleblog.com/ 2019/08/on-device-real-time-hand-tracking-with.html. Accessed 12 Jan 2023 6. Handtrack.js: tracking hand interactions in the browser using tensorflow.js and 3 lines of code, https://blog.tensorflow.org/2019/11/handtrackjs-tracking-handinteractions.html. Accessed 13 Jan 2023 7. Hand tracking and gesture control. https://www.manomotion.com/. Accessed 16 Jan 2023 8. Spranger, J., Buzatoiu, R., Polydoros, A., Nalpantidis, L., Boukas, E.: Humanmachine interface for remote training of robot tasks. In: 2018 IEEE International Conference on Imaging Systems and Techniques (IST), pp. 1–5. IEEE (2018) 9. Hand tracking gesture detection. https://www.youtube.com/watch? v=lBzwUKQ3tbw. Accessed 18 Oct 2022 10. Hand pose detection. https://developer.oculus.com/documentation/unity/unityisdk-hand-pose-detection/. Accessed 23 Jan 2023 11. Goldstone, W.: Unity Game Development Essentials. Packt Publishing Ltd. (2009) 12. Hillmann, C.: Comparing the gear VR, oculus go, and oculus quest. In: Hillmann, C. (ed.) Unreal for Mobile and Standalone VR: Create Professional VR Apps Without Coding, pp. 141–167. Springer, Heidelberg (2019). https://doi.org/10.1007/ 978-1-4842-4360-2_5 13. OctoXR. https://spectrexr.io/octo-xr. Accessed 19 Nov 2022 14. Dot product. https://mathinsight.org/dot_product. Accessed 10 Dec 2022

Transmission of Electromagnetic Waves Through a Clay Material Vanja Mandri´c(B)

, Slavko Rupˇci´c , Davor Vinko , and Domagoj Bilandžija

Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, Croatia [email protected]

Abstract. This paper deals with the simulation calculation of transmission (S21 ) and reflection (S22 ) parameters in a material parametrically based on clay (brick). The electromagnetic parameters of the clay that are the subject of the study are: relative permittivity εr , loss tangent tan δ and electrical conductivity σ, while the only structural parameter is the thickness d of the sample. The values of all considered parameters are within the range of data from available literature and values that could occur due to external causes (e. g. an increase in clay moisture leads to an increase in electrical conductivity). The simulation model was carried out with the software tool ANSYS HFSS and validated with a laboratory measurement model. The laboratory model consists of a circular waveguide with an inner diameter of 150 mm and a total length of 700 mm, which is excited by 52 mm long probes placed 50 mm from the last wall of the waveguide. The lower cut-off frequency of this measurement system is 1.173 GHz, so all results (simulation and measurement) are limited by this lower cut-off frequency of the waveguide. All simulation calculations were performed for three clay sample thicknesses: 18 mm, 36 mm and 54 mm. These calculations were made with the aim of determining brick composites that have high attenuation of electromagnetic EM waves (low transmission parameter and high reflection parameter) in order to reduce the strength of EM fields inside the object made of such bricks. The measured results were used to verify the simulation model. Keywords: Clay Brick · Electromagnetic Field · EM Wave Attenuation · Transmission Parameter · Reflection Parameter · Shielding Efficiency

1 Introduction Non-ionizing radiation (electromagnetic - EM) is becoming one of the most widespread environmental pollutants, and as the number of electrical devices increases, it is expected to continue to increase. However, along with this increase came awareness of the need to control levels and reduce EM radiation, especially in the space where people live. Some of the more significant research dealing with this topic are: 1997, W.C. Stone (NIST) research on the propagation of electromagnetic waves in building materials, including brick, eight different concrete mixes, and reinforced concrete, glass, and wood [1]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 75–85, 2024. https://doi.org/10.1007/978-3-031-51494-4_7

76

V. Mandri´c et al.

– In 2000, P. Pauli and D. Moldan conducted systematic tests of large building materials and special protective materials [2]. – NIST - The American National Institute of Standards and Technology and the University of the German Federal Armed Forces conducted tests in 2012 (concrete, brick, wood, drywall, plywood, glass and rebar) for frequencies covering the field of mobile communications 3G, 4G and LTE as well as frequencies of digital television, GPS and wireless smart meters [3]. – D. Micheli, et al. In 2014, measurements of the attenuation of electromagnetic waves (EM) in walls in (700 MHz–5 GHz) [4]. – M.R. Mahfouz, et al. 2014 measured RF attenuation in different types of walls (dry walls (up to 70 GHz), concrete (up to 3 GHz) and bricks (up to 10 GHz)) [5]. – ITU-R P.2040-1/2015 provides guidance on the effects of the electrical properties of building materials and structures on the propagation of radio waves above 100 MHz [6]. In the mentioned available literature, the measurements of the attenuation of EM waves through different types of building materials (concrete, brick, wood, glass, …) as well as the testing of the parameters of some of these materials are listed. The influence of the key electromagnetic parameters (permittivity, tangent loss and electrical conductivity) of these materials on the transmission and reflection, and thus on the attenuation of the EM wave when passing through them, has not been systematically analyzed, and this is the key contribution of this paper. In addition, the influence of the thickness of the material on the propagation of the EM wave as the most important structural parameter was processed by simulation. This paper deals with simulation calculations of EM wave transmission and reflection through clay materials of different electromagnetic and geometrical parameters. The second chapter deals with the definition of all electromagnetic parameters that are relevant for this research: relative permittivity, loss tangent and electrical conductivity of the material. The results of the simulation calculation of the coupling parameters (transmission and reflection) in terms of electromagnetic and structural and geometric parameters are given in Sect. 3. In addition, in the same chapter, the simulation calculations were verified by measurements. The last chapter brings the conclusions of this work and the possibility of continuing this research with other building materials.

2 Electromagnetic Parameters of the Shield Material An electromagnetic wave is transmitted through and reflected from both a conductive material, a dielectric, and a material that is a dielectric but has some electrical conductivity - a conductive dielectric (for which σ = 0 holds) - and in which the electromagnetic wave weakens as it propagates. Such a material is a true dielectric that has a certain moisture content that increases the electrical conductivity of the material, or in which certain additives are contained that increase the electrical conductivity. In such a material, the complex relative permittivity is defined as: ε = ε − jε = ε0 εr − jε0 εr

(1)

Transmission of Electromagnetic Waves Through a Clay Material

77

The real part of the complex permittivity ε denotes the stored electric field, while ε is the imaginary part of the complex permittivity ε and represents the measure of the loss factor of the amount of scattered energy (Eq. (1)). The material included in this research (clay) is not a perfect dielectric, so there are certain losses in such a material. These losses can be expressed as losses in the dielectric εd and losses caused by the electrical conductivity εc of the specified material: σ (2) εr = εd + εc = εd + ωε0 where: εr - the relative imaginary part of the ε ; εd - the relative dielectric component of the imaginary part of the ε ; εc - the relative conductive component of the imaginary part of the ε ; δ - dielectric loss angle and tan δ - loss tangent; σ - electrical conductivity; ω = 2πf - angular frequency of an EM wave and, ε0 - the vacuum permittivity. Loss tangens can be expressed by the equation: tgδ =

ε σ , and for εd  εc => tgδ   ε ε0 ω

(3)

The electrical conductivity of the material shown by the second term of the sum in Eq. (2) depends on the frequency and (it can be mainly caused by the presence of moisture in the material) and if this term is dominant in the sum (Eq. 3) then the approximate value of the loss tangent tgδ can be expressed as the ratio of the electrical conductivity σ and multiple of angular frequency ω and the vacuum permittivity ε0 (right part of Eq. (3)). Permittivity can be expressed through relative permittivity and loss tangent: ε = ε (1 − jtgδ)

(4)

The EM field can be expressed by a system of two equations - the equation of the electric and magnetic fields of that wave:   √  



j ωt−





  √  j ωt− ε (1−jtgδ)k0 r

E = E0 e

H = H0 e where: 

E0 – magnitude of electric field, V/m; 

H0 – magnitude of magnetic field, A/m; 

k0 – wave number in free space, m−1 ; ε – real part of dielectric permittivity; ω – angular frequency;  r – spatial distance, m; tgδ – loss tangent.

ε (1−jtgδ)k0 r

(5) (6)

78

V. Mandri´c et al.

3 Simulation Calculation of Coupling Parameters Through a Clay Block All simulation calculations in this paper were performed as full wave analysis using simulation software HFSS. The simulation calculations performed in this paper included the calculation of coupling parameters (S parameters) with special reference to the transmission parameters (S21 , S12 ) for the case of EM wave propagation through brick material. The whole simulation model is designed to describe as closely as possible the measurement system shown in Fig. 1. Figure 1 shows the simulation model used in these calculations.

Fig. 1. Simulation model of circular waveguide measurement system.

Although the reflection coefficients are not the focus of this paper, a simulation of the reflection (reflection coefficient S22 ) was made with the aim of considering its change for different changes in the sample parameters. The dimensions of the simulation model are as follows: a) b) c) d) e) f)

total set length 700 mm; inner diameter 150 mm; diameter of the Al ring 134 mm; length of excitations (Probe 1 and Probe 2) 52 mm; distance from the rear wall of the waveguide 50 mm and sample thickness (brick) 18–54 mm.

In addition to matching the dimensions of the simulation model with the measurement set, an important part of the entire simulation calculation is defining the electromagnetic parameters of the material from which the sample is made. The most important EM

Transmission of Electromagnetic Waves Through a Clay Material

79

parameters are relative permittivity (εr ), dielectric losses (tanδ) and electrical conductivity (σ ). In addition to these parameters, the final results of the simulation significantly depend on the thickness of the material. According to the available literature [7] in the 0.9 GHz–9 GHz frequency range the relative permittivity of brick (clay) material ranges from 3.30 to 4.62 for the real part and between 0.12 and 0.34 for the imaginary one. The electrical conductivity of the brick for the frequency range 0.9 GHz–9.0 GHz (according to the above measurements [7]) is in the range from 0.01 S/m–0.12 S/m which declares it as a dielectric material. The dielectric losses were calculated from the data of the real and imaginary parts of the permittivity and the electrical conductivity according to [7] and range from 0.037 to 0.041 (last column in Table 1). Table 1 contains the data used to define the frequency dependence of the relative dielectric constant and the loss tangent in the simulation calculation of coupling parameters. Table 1. Frequency dependence of relative permittivity and loss tangent of clay [7]. Frequency, GHz

Relative permittivity

Tangent loss

0.9

4.60

0.041

1.7

4.62

0.039

5.0

4.12

0.039

5.3

4.10

0.037

5.8

3.58

0.090

9.0

3.90

0.037

Typical density values of standard dry bricks values range from 1600–1915 kg/m3 . Depending on the type of brick material and structure [8]. The density of the sample has an effect on the final result, but within the limits of the brick density it does not significantly affect the S parameters. Since the ranges of values of the important EM parameters are within large limits, it was necessary to investigate their influence on the final results (S parameters). However, for more precise implementations of the simulation calculation, it was necessary to determine the densities of the samples used in the measurement and check whether they are in accordance with the data from the literature. The density of the samples was determined by precise measurements of the dimensions and mass of the samples. The densities of the clay samples (used in measuring the coupling parameters) range from 1050.2 kg/m3 to 1683.5 kg/m3 . Therefore, the simulations were performed in this density range. Although all structural and EM parameters have an impact on the transmission of EM energy through the brick material. Thus, the thickness of the sample is the most important of the dimensional parameters, and of the electromagnetic ones: relative permittivity, dielectric losses and electrical conductivity.

80

V. Mandri´c et al.

Simulation calculations were performed with the following parameters of the brick material are (Figs. 3, 4 and 5): – relative permittivity: 4.2 – bulk conductivity: 0 (0–10) S/m; – dielectric loss tangent: 0 (0–0.35); and dimensional - structural parameters of the brick sample are: – sample thickness: 18 mm, (18 mm, 36 mm and 54 mm). – mass density: 1600 kg/m3 . For Fig. 2, the parameter values are as follows: – – – – –

relative permittivity: freq. Variable (3.75–4.62); bulk conductivity: 0 S/m; dielectric loss tangent: 0.035); and dimensional parameters of the brick sample are: sample thickness: 36 mm; mass density: 1600 kg/m3 .

The results of the simulation calculation of coupling parameters (S22 and S21 ) with changes in the dielectric constant, sample thickness, loss tangent and electrical conductivity are shown in Fig. 2, 3, 4 and 5. An increase in the relative dielectric constant results in a partial decrease in the transmission parameter S21 (blue and red curves) at resonant frequencies, and this decrease is more significant in the sample with frequency-variable relative permittivity values in the reflection parameter S22 (green curve) (Fig. 2).

Fig. 2. Results of the simulation calculation of coupling parameters for clay material with variable value of relative permittivity: a) S21 , b) S22 .

Increasing the sample thickness results in a decrease in transmission (parameter S21 ) and an increase in reflection (parameter S22 ) as shown in Fig. 3. However, in some parts of the spectrum, this general conclusion has not been confirmed, but the most probable reason for this is constructive interference, such as in the 3 GHz to 3.5 GHz band. Furthermore, with increasing frequency the transmission of the EM wave (parameter

Transmission of Electromagnetic Waves Through a Clay Material

81

Fig. 3. Results of the simulation calculation of coupling parameters for clay material with variable value of thickness: a) S21 , b) S22 .

S21 ) decreases more strongly than at lower frequencies while the reflection (parameter S22 ) increases. As far as losses in the dielectric are concerned, they significantly affect the transmission and reflection, so that with the increase of the tanδ coefficient, the transmission decreases significantly, as well as reflection parameter. Thus, the transmission coefficient (S21 ) decreases in the range of 2 dB–30 dB in the analyzed frequency range (1.5 GHz–6 GHz), and the reflection coefficient (S22 ) of 1 dB– 12 dB. The tendency of decreasing transmission and reflection with increasing frequency is present in this case as well (Fig. 4). In the simulation calculation with variable electrical conductivity (Fig. 5) of the sample material, as expected, the transmission decreases significantly with increasing conductivity and this tendency is visible for all analyzed values of electrical conductivity.

Fig. 4. Results of the simulation calculation of coupling parameters for clay material with variable value of dielectric loss: a) S21 , b) S22 .

82

V. Mandri´c et al.

Thus, the value of the S21 parameter decreases by 2 dB–10 dB at a conductivity of 0.1 S/m, for 10 dB–40 dB at a conductivity of 1.0 S/m and for 40 dB–69 dB at a conductivity of 10 S/m. The reflection coefficients in the composite with increasing electrical conductivity show a slight decrease in values although their increase is to be expected. The reason could be insufficient electrical conductivity of the material for some visible increase in reflection. In the simulation calculation with variable electrical conductivity of the sample material, the reflection decreases with increasing conductivity and this tendency is more pronounced for higher frequencies (over 3.5 GHz). By adding supplements to the base clay in the production of bricks, which increase: electrical conductivity and dielectric losses in the material, it is possible to significantly reduce the transmission of EM waves through such material. This increases the attenuation of the EM wave and provides a space with a low level of EM field inside the building that would be made with such material.

Fig. 5. Results of the simulation calculation of coupling parameters for clay material with variable value of electric conductivity: a) S21 , b) S22

In order to verify the simulation model, the transmission parameter S21 was measured for the identical frequency range as in the simulation calculation from 1.5 GHz–6 GHz. Measurement and comparative simulation (Fig. 6) were performed with the following parameters of clay material are: – relative permittivity: 4.2 (assumption) – bulk conductivity: 0 S/m (assumption); – dielectric loss tangent: 0.037 (assumption); and dimensional - structural parameters of the brick sample are: – sample thickness: 20.5 mm; – 2 r = 150 mm; – mass density: 1683.5 kg/m3

Transmission of Electromagnetic Waves Through a Clay Material

83

Fig. 6. Measured and simulated results of the transmission parameter S21 of the clay sample in the frequency range from 1.5 GHz–6 GHz.

A comparison of the measured and simulated values shows a good matching of the values which suggests a good design of the simulation model with some deviation in the two segments of the frequency spectrum. In the ranges from 3.7 GHz–5.0 GHz and 4.0 GHz–4.70 GHz there is a deviation of the measured and simulated values, and the reason is in the probable incomplete compliance of the parameters of the clay material used in the simulation and measurement.

4 Conclusion This paper deals with the transmission and reflection of electromagnetic waves (EM) through clay blocks in the frequency range from 1.50 GHz–6.0 GHz. The paper presents the results of the simulation calculation of S21 and S22 parameters. The simulation was performed using the HFSS simulation software model. EM values were obtained from the available literature, and geometrically by measuring real models of clay blocks. The results of the simulation calculations show the following: 1. Geometrical parameter (thickness) – by increasing the sample thickness, the transmission parameter (S21 ) decreases in the entire frequency range (1.50–6.0 GHz), while the reflection parameter is not significantly affected by the sample thickness but they have a frequency-selective character. It is known that by increasing the thickness of the material through which the EM wave passes, reflection occurs on the two outer edges of the material, which affects the reflection of the material inside the material and the reflection at the entrance of the EM wave into the material, and thus the transmission. Depending

84

V. Mandri´c et al.

on the ratio of the thickness of the material and the wavelength of the EM wave in the material as well as at the entrance to the material, constructive and destructive interference occurs, which results in frequency-selective coefficients of reflection and transmission, which is visible on the simulation diagrams; 2. EM parameters (relative permittivity, loss tangent and electrical conductivity) – the change in relative permittivity values does not significantly affect the reflection and transmission parameters (for the observed range of relative permittivity values from 3.58–4.62) except for the model of frequency-variable values of relative permittivity and dielectric losses (for the reflection parameter S22 ), which is related to real data according to available literature. Although it is known from the literature that permittivity significantly affects the propagation of EM waves through the material, in the range of clay permittivity values from 3.58 to 4.62. This influence is not significant; – by increasing the loss tangent, the reflection and transmission parameters are significantly reduced, so that the biggest reduction compared to the zero value of the loss tangent is that for tan δ = 0.035 of −37 dB at the frequency 5.95 GHz. The loss tangent has a significant impact since it represents the dielectric losses in the material. The selected range of values is greater than the range of values related to clay (0.037 to 0.09), but the diagrams of the reflection and transmission coefficients show that even small changes in this parameter from, for example, 0 to 0.0035 show significant changes in the transmission parameter (max.: −12 dB at a frequency of 5.9 GHz) and reflection (max: −3.8 dB at a frequency of 5.9 GHz); – increasing the electrical conductivity of clay significantly reduces the transmission parameter, so compared to the zero value of electrical conductivity, the largest reduction in the transmission parameter is for σ = 10 S/m and is −74 dB at the frequency of 5.75 GHz. A decrease in the reflection parameter with an increase in electrical conductivity is noticeable at frequencies above 3.50 GHz. These calculations were made with the aim of determining brick composites that would have a high damping of EM waves (low transmission parameters and high reflection parameters), in order to reduce the strength of the EM fields inside the object made of such bricks. The measured results were used to verify the simulation model and show a good agreement between the simulation and measurement results. The simulation model used should be refined in the future by introducing frequency dependence and other EM parameters of clay materials and by creating new mixtures and their models for clay mixtures that could result in higher amounts of EM wave transmission reduction through such materials.

References 1. Stone, W.C.: Electromagnetic signal attenuation in construction materials. NISTIR 6055, NIST construction automation program report no. 3. Building and Fire Research Laboratory National Institute of Standards and Technology Gaithersburg, Maryland (1997) 2. Pauli, P., Moldan, D.: Reduction and shielding of RF and microwaves: construction materials, screens, wainscots and tissues. University of the German Federal Armed Forces, Iphofen, Germany (2000)

Transmission of Electromagnetic Waves Through a Clay Material

85

3. NIST, EMF shielding by building materials, attenuation of microwave band electromagnetic fields by common building materials. U.S. National Institute of Standards and Technology (NIST) (2012). (Updated 2017) 4. Micheli, D., Santoni, F., Delfini, A., Marchetti, M.: Measurement of electromagnetic field attenuation by building walls in the mobile phone and satellite navigation frequency bands. IEEE Antennas Wirel. Propag. Lett. 14, 698–702 (2014) 5. Mahfouz, M.R., Fathy, A., Badawi, A.: See-through-wall imaging using ultra wideband pulse systems. Conference Paper (2005) 6. Recommendation ITU-R P.2040-1. Effects of building materials and structures on radiowave propagation above about 100 MHz, P Series Radiowave propagation (2015). http://www.itu. int/ITU-R/go/patents/en 7. Choroszucho, A., Butrylo, B., Steckiewicz, A., Stankiewicz, J.M.: Determination of the effective electromagnetic parameters of complex building materials for numerical analysis of wireless transmission networks. Electronic 9(10), 1569 (2020) 8. Hlavacova, Z.: Electrical properties of some building materials. Research and Teaching of Physics in the Context of University Education Nitra (2007)

Opening Doors and Drawers by a UR5 Robot with Force Control Jana Duki´c(B) , Lukrecia Vuli´c, Valentin Šimundi´c, Petra Peji´c, and Robert Cupec Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, J. J. Strossmayer University of Osijek, Osijek, Croatia [email protected]

Abstract. Robots capable of automatically opening and closing articulated objects such as doors and drawers have broad potential applications in households, healthcare facilities, and industrial environments. Drawers can be opened and closed by pulling or pushing along the axis of translation, while doors can be opened by rotating around the axis of rotation. When a robot’s motion path deviates from the ideal trajectory required for opening and closing, its action can cause various damages due to excessive lateral forces. In this paper, we address the problem of force control in robotic manipulation of doors and drawers. The goal of force control is to minimize lateral forces when opening and closing doors and drawers. At the same time, position control is performed in the direction of the ideal trajectory. A force-torque sensor attached to a robot can monitor all forces in 3D space. We conduct experiments on opening and closing the doors and drawers using a UR5 robot with and without force control, and analyze the force response while monitoring the success of the actions performed. With this study, we provide insight into the usefulness of force control in opening and closing doors and drawers by a robot. Keywords: robot manipulation · force control · trajectory · articulated objects · force-torque sensor

1 Introduction In recent years, robotics has witnessed great advancements in various domains, particularly in the field of object manipulation. Robots are increasingly being deployed in real-world scenarios, where they are tasked with performing complex operations such as opening and closing doors, and accessing the contents of drawers. Accomplishing these tasks with precision and efficiency requires the incorporation of force control mechanisms into robotic systems. Articulated objects, such as doors and drawers, present unique challenges for robots due to their inherent mechanical complexity. Unlike rigid objects, articulated objects possess joints, hinges, and varying degrees of freedom, making their manipulation a non-trivial task. A mere application of motion control algorithms may lead to imprecise movements, excessive force exertion, and even damage to the object or the robot itself. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 86–96, 2024. https://doi.org/10.1007/978-3-031-51494-4_8

Opening Doors and Drawers by a UR5 Robot with Force Control

87

Force control, on the other hand, empowers robots with the ability to adapt to dynamic environments and accurately interact with articulated objects. By sensing and regulating the forces applied during manipulation, robots can adapt their actions in response to changes in contact forces, object compliance, and external disturbances. This level of adaptability enables robots to perform delicate operations while maintaining safety and precision. The contribution of this paper is to thoroughly examine and emphasize the necessity of incorporating force control in robotic manipulation of articulated objects, focusing specifically on doors and drawers. Through a series of experiments and analyses with a UR5 robot arm, we aim to investigate the advantages of force control in achieving successful interactions with such objects. The paper is organized as follows: Sect. 2 presents trajectories for opening and closing doors and drawers. Position/force control is explained in Sect. 3. Section 4 explains the experiments and performs an experimental analysis by monitoring the force values during opening and closing doors and drawers with and without using the force control. Section 5 concludes the paper. 1.1 Related Research Hybrid force-position control has emerged as a promising approach for enhancing the capabilities of robots in various applications, including door opening. The existing literature reveals several key trends and techniques in this field. The incorporation of force and force-torque sensors in robot wrists has proven beneficial for door opening tasks [1–9]. These sensors provide valuable tactile feedback, enabling robots to perceive and respond to external forces exerted during the manipulation process, ensuring safer and more precise door opening operations. The hybrid force-position control strategies have been investigated, although not exclusively for door opening [10]. These methods integrate both force and motion cues to achieve precise manipulation tasks, showcasing their potential for enhancing robot performance in door opening scenarios. Furthermore, adaptive force-velocity control techniques have been proposed to address the challenge of opening unknown doors [1, 3]. These approaches enable robots to autonomously adapt their force and velocity based on real-time feedback, allowing them to efficiently manipulate doors with varying characteristics. Another significant aspect is the application of adaptive impedance or model predictive control in tasks such as grinding and polishing [2], door opening and object lifting [5]. These studies offer valuable insights into the potential use of these control methodologies in manipulating objects with complex dynamics, including doors. Moreover, the integration of vision and force control has been investigated, presenting an interesting avenue for door opening [6–9]. Combining visual perception with force feedback allows robots to exploit both visual cues and physical interaction forces to optimize their door manipulation strategies. Finally, research has explored the learning of force-relevant skills from human demonstrations [4, 10]. By observing and imitating human operators, robots can acquire intuitive force-control strategies for interacting with doors, improving their adaptability and dexterity.

88

J. Duki´c et al.

Our paper contributes to this field by presenting an experimental evaluation using the UR5 robot manipulator. Through our experiments, we demonstrate that force control is essential for successfully performing door and drawer opening tasks, validating the importance of integrating force-based strategies in robot manipulation for these specific applications.

2 Trajectory The trajectory of an object is defined as a path to follow to reach the desired position. In order for the robot to start the action of opening doors or drawers, it must know in advance the required trajectory of the articulated object in question. When opening doors and drawers, the desired position is reached when the door or drawer is fully open. Both doors and drawers have limited freedom of movement and have different trajectories. Therefore, two different movements can be distinguished: doors can be opened and closed by rotating around the axis of rotation, while the drawers can be opened and closed by pulling or pushing along the axis of translation. 2.1 Door Opening Trajectory Let us consider the reference frames shown in Fig. 1. The pivot point of the door represents the origin of the closed door reference frame, SA , whose z-axis, zA , is the axis of rotation. The reference frame of the gripper is represented by SG , and the robot’s base reference frame is represented by SB . The tool reference frame ST is assigned to the last link of the robot arm to which the gripper is attached. This reference frame is not shown in Fig. 1. The angle θ represents the door opening angle and a represents the distance between the axis of rotation and the position of a handle. When the door is opened, initial reference frames SA , SG and ST change the pose and are finally represented by reference frames SA , SG and ST  . Door opening trajectory consists of points represented by the origin of the reference frame ST , as the starting point, and origins of reference frames ST  , for increasing values of the opening angles θ . Given an initial pose of the robot tool with respect to the robot’s base, defined by transformation matrix TTB , the tool pose after opening the door by the angle θ is calculated by TTB = TTB TGT TGA

−1 A A T  −1 TA TG TG

(1)

Since each of the reference frames SA , SG and ST rotate by the same angle, their relative position and rotation do not change, therefore 

TGT = TGT 

TGA = TGA

(2) (3)

Opening Doors and Drawers by a UR5 Robot with Force Control

89

Fig. 1. Door opening trajectory.

Relative position and rotation of reference frame SG in relation to SA is represented by ⎡

⎤ 0 0 1 0 ⎢ −1 0 0 −a ⎥ ⎥ TGA = ⎢ ⎣ 0 −1 0 0 ⎦ 0 0 0 1

(4)

The orientation of the gripper with respect to the door reference frame SA is defined according to a known suitable orientation of the gripper fingers with respect to the door handle. Reference frame SA rotates along its z-axis as noted by  Rz 0 TAA = (5) 0 1 where RZ represents rotation matrix about z-axis. The transformation matrix TTB is defined by manually positioning the robot in a suitable position for grasping the handle. Given a known orientation of the gripper with respect to ST , and value b representing the distance between the tip of the gripper and the origin of ST , the pose of the gripper with respect to ST can be defined by ⎡

⎤ 0 ⎢ RT 0 ⎥ G ⎥ TGT = ⎢ ⎣ b⎦ 0 1

(6)

Based on the matrices given above, it is possible to calculate the next point on the trajectory using Eq. (1). The calculation of each point is recursively based on the previous point on the trajectory.

90

J. Duki´c et al.

2.2 Drawer Opening Trajectory The trajectory of drawer opening and the associated reference frames are shown in Fig. 2. It can be seen that this trajectory has only one degree of freedom, namely translation. The reference frame of the drawer is represented by SA , which is located in the contact point between the surface of the drawer and the robot gripper. In each iteration, the drawer is opened by the distance d, while a2 represents the distance between the reference frames SA and SG, which represents the gripper.

Fig. 2. Drawer opening trajectory.

Due to the different placement of the handle on the drawer compared to the placement on the door, when grasping the drawer, the matrix TGA is calculated as follows ⎡

⎤ 0 0 1 −a2 ⎢ −1 0 0 0 ⎥ ⎥ TGA = ⎢ ⎣ 0 −1 0 0 ⎦ 0 0 0 1

(7)

Considering that coordinate systems SA and SA are not rotated like in the case of door opening, but translated, their relation is presented as follows. Due to the rectilinear motion of drawer opening, rotation between SA and SA coordinate system is represented as a unit matrix where the translation vector represents the length by which the drawer is opened. ⎡

TAA

⎤ 1 0 0 −d ⎢0 1 0 0 ⎥ ⎥ =⎢ ⎣0 0 1 0 ⎦ 000 0

(8)

Opening Doors and Drawers by a UR5 Robot with Force Control

91

The trajectory model represented by the Eq. (1), is further computed as explained in the previous section.

3 Position/Force Control Force control in robotics is a technique for regulating the force applied in performing a specific task. The force control problem is considered a natural progression of the motion control problem. The focus of model control is on regulating the motion of the robot manipulator, while the focus of force control is to use the feedback from the force/torque sensor to control the interaction between the robot manipulator and an environment that does not have a well-defined structure [11]. Combining these two strategies creates a hybrid position/force control which is used to solve the problem of opening and closing doors and drawers [11]. During the execution of a task in which a robot is in contact with objects in its environment, the robot’s movement can be constrained in certain directions. The position/force control strategy works by controlling the position in the unconstrained directions, while controlling the force in the constrained directions [11]. When opening and closing doors and drawers, the unconstrained direction is the direction in which the door or drawer opens, while the other directions are constrained. In this case, the target value of the force used in the constrained directions is zero, since the door or drawer does not open or close in that direction. For any operation of opening or closing a door or a drawer, there is an ideal path on which the lateral forces, i.e., the forces perpendicular to the direction of opening or closing, are zero. In practice, however, this path is not known. Instead, an approximation of this path is calculated using available information about the opening or closing direction obtained from a perception sensor with limited accuracy. The computation of such a path is shown in Sect. 2. We execute a trajectory that moves the robot tool along this approximate path using force control. This ensures that any deviation of the calculated trajectory from an ideal path in constrained directions is corrected by the force control, which regulates the forces in these directions to zero. The main objective of position/force control when opening and closing doors and drawers is to ensure that these movements can be performed without damaging the robot manipulator or the furniture because too much force is applied due to an inaccurately calculated trajectory. The robot manipulator using this type of control strategy should corrects its position in real time while performing a task, such that its motion in the unconstrained direction is rigid and follows the position specified in the trajectory, while at the same time it is “slack” in the constrained directions and corrects its position to ensure that the force value in these directions approaches zero. A visualization of the correction of the door opening trajectory by force control is given in Fig. 3. The black colored trajectory represents the actual trajectory of a door. The purple line is the planned trajectory based on data entered using robot vision or manually. The green line is the door at time t when the trajectory is executed. Point A represents the tool center point - TCP at time k. Let us assume that the force control is set to be controlled by position in the z-direction of the tool and by force in the x and y lateral directions, where the desired force is 0 N. The point A where TCP is actually located corresponds to the point B on the planned trajectory where TCP should be located at

92

J. Duki´c et al.

time k, but there is a deviation given by the parameter deviation range allowed to be able to control the force in x and y directions. The tolerances given by the deviation range parameter are shown with pink lines. Point C is the point on the planned trajectory where TCP should be at time k + 1. The blue vector represents the z-axis of the tool at time k. Point D has the same z-coordinate as point C, but there the lateral forces are equal to zero (the blue line is perpendicular to the z-axis of the tool), so this is exactly the point where TCP should arrive at time k + 1. In the way described, TCP should move on a black trajectory where the lateral forces are equal to zero.

Fig. 3. Visualization of door opening trajectory correction by force control.

4 Experimental Evaluation Position/force control experiments were performed using a UR5 robotic arm with an attached FT 300-S force-torque sensor and Robotiq’s adaptive 3-finger robotic gripper, as shown in Fig. 4. To find out if the UR5 robot can successfully open doors and drawers, two experiments were conducted. In the first experiment the robot opened a drawer, in the second a door. In each experiment, the robot was manually placed in front of the drawer handle or door handle. Then the algorithm was started to close the gripper and calculate the trajectory to open/close the door and drawer. The trajectory models are explained in Sect. 2 and implemented in the program. Robot movement by these trajectories is performed in the experiments. Each experiment was performed first with and then without force control by using Force mode, a Universal Robots operating mode that allows the robot to operate in a force-dependent manner. Force mode provides the ability to regulate forces and torques along each of the three axes, allowing specific values of these variables to be set. Force/torque control is applied to all manipulation commands invoked following Force mode. This mode is particularly useful in tasks that require precise force control, as it allows the robot to interact with its environment in a controlled and adaptive manner. During the experiments, the force values were monitored and sampled at a frequency of 100 Hz.

Opening Doors and Drawers by a UR5 Robot with Force Control

93

The first experiment consisted of opening and closing the drawer shown in Fig. 4. Figure 5 shows the force values measured by the force-torque sensor at the TCP for opening and closing the drawer with and without Force mode. If we compare these two processes, we find that when force control is activated, the result of the force in the x-direction and in the y-direction deviates around zero. In the x-direction force ranges mostly from −5 to 5 N and in the y-direction it ranges from −3 to 7 N. Measured forces in the z-direction have smaller values if force control is activated in x and y direction. Based the values of the force in z-direction it is possible to distinguish the exact moment of direction change, from pulling motion to pushing motion. The movement when the drawer opens is recorded in the first 5 [s] and the closing of the drawer is seen after 5 [s], with the force changing the direction of the effect. However, with the force control disabled, the robot managed to open the drawer due to the precise positioning of the gripper relative to the handle and due to the built-in motion function, where the motion of the robot is linear in the Cartesian coordinate system, which means that the robot’s TCP maintains the rectilinear motion along the z-axis. Robot’s ability to pull and push the drawer was hindered due to the restricted freedom of movement and the inability to adjust the trajectory when force control was not activated. At some moments, it was visible to the human eye that the robot was pulling the drawer up and down and even left and right, which resulted in slight movements of complete cabinet. This can be seen in the force curve, where the forces increase by 10 N in the x-direction, which means that more force was required to open the drawer in this case.

a)

b)

Fig. 4. The experimental setup of a UR5 robot opening a drawer: a) Initial state of drawer opening, b) Final state of drawer opening.

The second experiment consisted of opening and closing the door, as shown in Fig. 6. First, force control was activated, which resulted in the robot managing to open and close the door. The forces obtained are shown in Fig. 7. In this case, the robot needed more time to open and close the door than to open and close the drawer, a total of 35 s. In the first 15 s the robot opened the door and after the 15th second it closed it again. From Fig. 7, it can be seen that the force values of the first samples are quite high, which is normal because the handle is gripped with a lot of force to keep it in the gripper during the opening and closing of the door. In the x-direction, the force ranges from 3 N to 12 N when opening the door and from 7 N to −5 N when closing it. In the y-direction, the

94

J. Duki´c et al.

a)

b)

c)

Fig. 5. Comparison of results – time series plots of the force measured by the force-torque sensor while opening and closing the drawer with (blue) and without (red) using the Force mode: a) Force in X-direction, b) Force in Y-direction, c) Force in Z-direction.

force ranges from 0 N to −20 N to open the door and from −20 N to 10 N to close the door. In the z-direction, the forces are between −6 N and 4 N. These large differences between the start and end forces indicate that the robot has adjusted its position to require the lowest possible force to open and close the door. Then, the force control was disabled, which resulted in the robot not opening the door. The second experiment was repeated 20 times, and each time the robot’s movement ended with Protective stop. Protective stop is a built-in feature of the UR5 controller that puts the robot in a protective state when maximum forces are reached. In this second experiment, it becomes clear how important force control is. Without this feature, the robot would not be able to adjust its position to stay within safe force limits and still open the door. Regardless of whether the Protective stop was enabled or not, robots that do not use force control can easily damage articulated objects, themselves or the environment. For this reason, force control is suggested as the key to safe handling of the above objects.

a)

b)

Fig. 6. The experimental setup of a UR5 robot opening a door: a) Initial state of door opening, b) Final state of door opening.

Opening Doors and Drawers by a UR5 Robot with Force Control

a)

b)

95

c)

Fig. 7. The dependence of force on duration measured by the force-torque sensor while opening and closing the door when using the Force mode: a) Force values in X-direction, b) Force values in Y-direction, c) Force values in Z-direction.

5 Conclusion This paper deals with the experimental evaluation of opening drawers and doors with a UR5 robotic arm, with and without the use of force control. While performing the experiments, force values were monitored in the x, y, and z-directions of the robot’s tool center point (TCP). Based on the conducted experiments, it can be concluded that the robot demonstrated success in opening the drawer, both with and without using the force control. Nonetheless, when force control was activated, the robot executed a visibly more elegant action in opening and closing the drawer. On the contrary, the more complex task of opening and closing the door necessitated force control, even for small opening angles. The two experiments conducted demonstrate the advantages of force control for manipulation with articulated objects, both for the safety of the robot and the environment, and for the success of the opening and closing operations of the articulated objects. In future work, position/force control should be integrated into a vision-based robotic system. The complete process of autonomous opening of doors and drawers should consist of detection of such furniture in the scene, calculation of kinematic parameters and trajectory models of opening of their articulated parts, navigation of a robotic arm to the furniture handle, and finally execution of the task of opening or closing with activated force control.

References 1. Karayiannidis, Y., Smith, C., Viña, F.E., Ogren, P., Kragic, D.:“Open sesame!” adaptive force/velocity control for opening unknown doors. In: IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura-Algarve, Portugal, pp. 4040–4047 (2012). https:// doi.org/10.1109/IROS.2012.6385835 2. Zhou, H., Ma, S., Wang, G., Deng, Y., Liu, Z.: A hybrid control strategy for grinding and polishing robot based on adaptive impedance control. Adv. Mech. Eng. 13(3), 168781402110040 (2021). https://doi.org/10.1177/16878140211004034

96

J. Duki´c et al.

3. Karayiannidis, Y., Smith, C., Barrientos, F.E.V., Ögren, P., Kragic, D.: An adaptive control approach for opening doors and drawers under uncertainties. IEEE Trans. Rob. 32(1), 161–175 (2016). https://doi.org/10.1109/TRO.2015.2506154 4. Gao, X., Ling, J., Xiao, X., Li, M.: Learning force-relevant skills from human demonstration. Complexity 2019, 1–11 (2019). https://doi.org/10.1155/2019/5262859 5. Minniti, M.V., Grandia, R., Fah, K., Farshidian, F., Hutter, M.: Model predictive robotenvironment interaction control for mobile manipulation tasks. In: 2021 IEEE International Conference on Robotics and Automation (ICRA), Xi’an, China, pp. 1651–1657. IEEE (2021). https://doi.org/10.1109/ICRA48506.2021.9562066. Accessed 7 July 2023 6. Schmid, A.J., Gorges, N., Goger, D., Worn, H.: Opening a door with a humanoid robot using multi-sensory tactile feedback. In: 2008 IEEE International Conference on Robotics and Automation, Pasadena, CA, USA, pp. 285–291. IEEE (2008). https://doi.org/10.1109/ ROBOT.2008.4543222. Accessed 7 July 2023 7. Prats, M., Martinet, P., Del Pobil, A.P., Lee, S.: Robotic execution of everyday tasks by means of external vision/force control. Intell. Serv. Robot. 1(3), 253–266 (2008). https://doi.org/10. 1007/s11370-007-0008-x 8. Arduengo, M., Torras, C., Sentis, L.: Robust and adaptive door operation with a mobile robot. Intell. Serv. Robot. 14(3), 409–425 (2021). https://doi.org/10.1007/s11370-021-00366-7 9. Stuede, M., Nuelle, K., Tappe, S., Ortmaier, T.: Door opening and traversal with an industrial cartesian impedance controlled mobile robot. In: 2019 International Conference on Robotics and Automation (ICRA), Montreal, QC, Canada, pp. 966–972. IEEE (2019). https://doi.org/ 10.1109/ICRA.2019.8793866. Accessed 7 July 2023 10. Wang, N., Chen, C., Nuovo, A.D.: A framework of hybrid force/motion skills learning for robots. IEEE Trans. Cogn. Dev. Syst. 13(1), 162–170 (2021). https://doi.org/10.1109/TCDS. 2020.2968056 11. Siciliano, B., Villani, L.: Introduction. In: Siciliano, B., Villani, L. (eds.) Robot Force Control. The Springer International Series in Engineering and Computer Science, vol. 540, pp. 1–6. Springer, Boston (1999). https://doi.org/10.1007/978-1-4615-4431-9_1

Maintaining Mobile Communication in Distress and Emergency Situations Tomislav Bari´c

and Hrvoje Glavaš(B)

Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Osijek, Croatia [email protected]

Abstract. This paper provides an overview of the technical aspects of the application of amateur radio stations in maintaining mobile communication in distress and emergency situations. Ultra-high frequencies (UHF) were selected for radio communication testing. Testing was performed at a frequency of 446 MHz, which is covered by the PMR446 (Private Mobile Radio) standard. Handheld radios were used to test communication in urban conditions. Given the high attenuation of electromagnetic waves in urban areas due to concrete and steel reinforced concrete buildings, special attention is paid to testing communication in urban areas. The quality and range of communication using non-directional and directional antennas as well as with simplex repeaters are presented in the paper. The obtained measurement results of quality and range of communication in urban conditions are analyzed and interpreted. General conclusions are given on the capabilities of amateur radio handheld stations and additional equipment such as antennas and simplex repeaters for maintaining communication in distress and emergency situations in urban conditions. Keywords: Amateur radio · Emergency situations · Distress situations · Directional antenna · Mobile communications

1 Introduction Unlike commercial communication systems, amateur radio has inherent properties that make it an independent and self-sufficient way of communication even when other communication systems fail. Amateur radio is dispersed throughout a community and locations and without a strict hierarchy, and without reliance on terrestrial fixed or mobile infrastructure. Therefore, such a system cannot be overloaded as it is possible to overload a cellular telephone system that relies on cellular telephone sites that can be overloaded or damaged. Radio amateurs are very skilled and experienced in the use of radio equipment and electronics. If there is a need for that, they can improvise antennas and power sources (car batteries, solar power, generators) in a very short time. At the same time, radio amateurs are extremely familiar with the local circumstances of radio wave propagation in the places where they live and the surrounding area. They are also aware of the capabilities of their radio equipment, and the range and quality of © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 97–112, 2024. https://doi.org/10.1007/978-3-031-51494-4_9

98

T. Bari´c and H. Glavaš

communication they can achieve. All of the above allows them to easily adapt to new circumstances in the event of an emergency. The ability of radio amateurs to adapt to distress circumstances to quickly establish communication when conventional systems are overwhelmed, collapsed or non-existent is well documented and recognized by the states administration [1–4]. The documented abilities of radio amateurs in establishing alternative communications relate to the use of mobile and stationary radio devices. With such radio devices, it is relatively easy to communicate at distances of tens and hundreds of kilometers. Instead of analyzing the capabilities of communication of such devices in urban conditions, the paper will analyze significantly weaker (in terms of output RF power) devices, i.e. handheld radios. In order to determine the capabilities of average amateur radio equipment, more precisely, handheld radio stations used by radio amateurs, for testing we chose a handheld radio that belongs to the middle price range. We performed the testing in Osijek, a town located in eastern Croatia. Depending on the period of construction, the city of Osijek has settlements with a small share of steel reinforced concrete buildings and settlements with a large share of steel reinforced concrete buildings. This allowed us to test the propagation of electromagnetic waves in different environments. The quality and range of communication using non-directional and directional antennas as well as with simplex repeaters are presented in the paper. The obtained measurement results of quality and range of communication in urban conditions are analyzed and interpreted. General conclusions are given on the capabilities of amateur radio handheld stations and additional equipment such as antennas and simplex repeaters for maintaining communication in distress and emergency situations in urban conditions.

2 Radio Amateurs in Distress and Emergency Situations Natural disasters such as fires, floods, hurricanes, tsunamis, huge snowdrifts, earthquakes, are regularly accompanied to some extent by the collapse of technical infrastructure. Overload or collapse of telephone lines, fiber optic cables and the global system for mobile communications (GSM), apart from causing coordination difficulties between people and services in the affected area, also has a psychological effect on them. Throughout the history of amateur radio, there are well-documented cases of active participation of radio amateurs in establishing communication in distress and emergency situations. Radio amateurs took part in reporting the situation in the affected area [5–7]. They also reported to the rescue services about newly created dangerous areas and about the number of injured people in the affected areas [5–7]. They also participated in the search for lost and isolated individuals [5–7]. In all these situations radio amateurs used: stationary radios, mobile radios in vehicles and handheld radios. These three classes of radio devices differ drastically in terms of communication range, communication quality and the number of devices available in the affected area. Stationary radio devices have the highest output RF power, and it is typical for their use to use high-quality, often directional antennas. Because of the above, they enable large communication distances, of the order of hundreds of kilometers. However, the expected number of such devices owned by radio amateurs in the affected area is very small. Mobile radio devices have a lower output RF power compared to stationary ones. They are used much more often among radio amateurs, and are often used as a replacement for stationary radio devices.

Maintaining Mobile Communication

99

Depending on the radio bands, output power and type of antenna, they enable communication distances from a few dozen kilometers to up to a hundred kilometers. Given that such radio devices are very widespread among radio amateurs, a considerable number of such devices can be expected in the affected area. Although there are many ways in which radio amateurs can organize themselves in establishing an alternative mobile communication network using handheld radio stations, in this paper some of the many possibilities has been selected and analyzed.

3 Recent Experiences, Earthquake in Petrinja Although most of the territory of the Republic of Croatia is excellently covered by the mobile cellular data network, the importance of alternative mobile communications has not diminished. There are several reasons why the existence of alternative mobile communications is important. Each of the reasons starts from the fact that the mobile cellular data network is not 100% available during the year, especially in exceptional situations. Such exceptional situations are power outages of the cellular data network, software upgrades of the mobile cellular data network, system outages due to malicious software (viruses), and system outages due to physical damages of the mobile cellular data network. The importance of alternative mobile communication is especially important in cases of various disasters such as floods, fires and earthquakes. To better understand the above, let’s look at a more famous and documented example from recent history. On 29 December 2020 at 12:20 PM (CET), an earthquake of magnitude 6.4 Mw (The moment magnitude scale) i.e. 6.2 ML (Local magnitude scale-Richter scale) hit central Croatia, with an epicenter located roughly 3 km west-southwest of Petrinja. Petrinja is a town in central Croatia near Sisak in the historic region of Banovina (Coordinates: 45°26 26 N 16°16 42 E) Due to the collapse of a large number of buildings, electrical connections were broken, which then pulled the poles of low-voltage (0.4 kV) electrical networks. Although the entire power system in Petrinja and surrounding areas was severely damaged by the earthquake, the most significant damage occurred at distribution transformer stations and the distribution electrical network. To illustrate the devastating effect of the earthquake, out of a total 243 tower-type substations, 69% needed to be completely replaced after the earthquake [8]. As the earthquake was felt in the wider area of Croatia, initially, using the mobile network, citizens tried to find out where the earthquake occurred and what the consequences were. Due to a large number of simultaneous calls and use of mobile internet, there was a short-term congestion of the network in most of Croatia. After about twenty minutes (experience of the authors of the article), the situation stabilized in most of Croatia. Severe interruptions, disturbances and congestions continued in the area of Petrinja, Sisak and Glina for several days. The reason why the interruptions, disturbances and congestions in communication in the area of Petrinja have continued is the following. The collapse of the buildings and towers where the radio base stations were installed led to the destruction of the base stations and the complete collapse of the mobile phone network in Petrinja. Due to the collapse of buildings, many power distribution cables and fiber optic cables were broken or damaged. The combined effect of a long-term power outage with broken optical cables led to the collapse of a certain part of the public switched telephone network (PSTN). Termination of terrestrial

100

T. Bari´c and H. Glavaš

and radio communication made it difficult to find and provide assistance to earthquake victims. The event opened up the issue of availability and reliability of the conventional communication network in distress and emergency situations.

4 Radio Amateur Handheld Radio Stations There are many factors and parameters on the basis of which hand-held radio stations can be categorized. Some of them are: transmitter power, receiver sensitivity, receiver selectivity, audio quality, audio noise cancellation, battery life, accessories, antenna quality, water resistance, shock resistance, the number of channels that can be memorized, visibility of screen contents in the sun and many others. Despite the very uniform data declared by the manufacturer on the transmitter power and receiver sensitivity and selectivity, in practice there is a significant difference in the range and quality of communication between them. Therefore, we recommend to interested individuals, when buying radio stations to consult on the quality of individual devices in amateur radio communities. The price of the most popular handheld radio amateur radio stations ranges from 50 to 250 euros. Without privileging any of manufacturer, we will list some of the more popular (well-known) handheld radios in alphabetical order: Alinco, Baofeng, Icom, Kenwood, Midland, Motorola, QuanSheng, Retevis, Wouxun, Yaesu. 4.1 Antennas for Handheld Radio Stations Hand-held radios have a low output RF power (up to 10 W), so the quality and range of communication depends greatly on the antennas used. Today, most manufacturers of hand-held radios supply at least one antenna together with the radio device. Most often, these are short monopole antennas, so-called rubber ducky antennas (also called stubby antennas). For amateur radio frequencies, the voltage standing wave ratio (VSWR) of such antennas is typically below 2. Although the VSWR [9] of such antennas is acceptable (Table.1. [10]), the transmission/reception quality of such antennas lags behind regular whip antennas. Whip antennas are longer compared to stubby antennas (Fig. 1.), and have higher gain and better (lower) VSWR (typically below 1.5). Table 1. Voltage standing wave ratio and reflected power. VSWR

1.0

1.2

1.4

1.6

1.8

2.0

Reflected Power (%)

0.0

0.8

2.8

5.3

8.2

11.1

Maintaining Mobile Communication

101

The combination of higher gain and better (lower) VSWR that whip antennas have compared to stubby antennas is the reason for noticeably better communication quality if whip antennas are used instead of stubby antennas.

Fig. 1. Handheld radio devices with supplied rubber ducky antennas, i.e. stubby antennas: a) Baofeng UV-5R, b) Wouxun KG-UV9T, c) Retevis RT83. Whip antennas for handheld radio devices: d) Midland G9, e) Retevis RHD-771, f) Diamond SRH 771.

In accordance with the above, for the purposes of this paper, i.e. determining the quality and range of communication in urban conditions, we used a whip antenna. We used a whip antenna manufactured by Retevis, model RHD-771, we tested that antenna. The testing of the basic electrical parameters was carried out at a frequency of 446 MHz, i.e. at the frequency at which measurements of the quality and range of communication in urban conditions were carried out. Basic electrical parameters and VSWR vs frequency of the antenna were measured using the PS100 RF vector antenna analyzer meter [11]. Screen shots of the vector analyzer during the measurement are shown in Fig. 2. At the frequency of 446 MHz, the VSWR of the antenna is 1.2. According to Table 1, the percentage of reflected power is approximately 0.8%.

102

T. Bari´c and H. Glavaš

Fig. 2. Basic electrical parameters and VSWR vs frequency of flexible whip antenna for model: Retevis RHD-771.

4.2 Directional Antennas Although radio amateurs like to experiment with self-construction of various types of directional antennas, at very high frequencies (VHF) and ultra high frequencies (UHF) the use of two types of directional antennas prevails. These are Yagi antennas and their subtypes (e.g. HB9CV) [12], and Quad antennas and their subtypes (e.g. BiQuad) [12]. For the purposes of this paper, a vertically stacked antenna composed of two identical Quad antennas, each with four elements, placed one above the other vertically, was used (see Fig. 3). This antenna was made by the authors of the article and is presented in detail in [13]. The electromagnetic characteristics of the antenna are simulated in the 4nec2 software [14]. The radiation diagram obtained by simulation is shown in Fig. 3. According to the simulation results (see Fig. 3), the antenna has 13 dBi gain. The result is expected because each individual quad antenna consisting of four elements has a theoretical gain of 10 dBi. By vertical stacking of antennas, a vertical narrowing of the radiation beam is achieved with an unchanged horizontal radiation beam compared to individual antennas. Theoretically, the additional gain of 3 dBi compared to individual antennas originates from the vertical narrowing of the radiation beam. A narrow vertical radiation beam and a relatively wide horizontal radiation beam is suitable in situations where one of the operators is mobile and its exact location is not exactly known. This is one of the reasons why we chose this antenna for testing communication in distress and emergency situations. The impedance matching on this antenna was achieved by using a coaxial cable [13]. Two meters of RG58 coaxial cable is used to connect the antenna and radio device. Taking into account the attenuation in the coaxial cables for impedance matching and for connecting the antenna to the radio devices, approximately one dBi is lost. Considering the losses in the coaxial cable, the antenna has an effective gain of 12 dBi [13]. Basic electrical parameters and VSWR vs frequency of the antenna were measured using the PS100 RF vector antenna analyzer meter [11]. Screen shots of the vector analyzer during the measurement are shown in Fig. 4. At the frequency of 446 MHz, the VSWR of the antenna is 1.14. According to Table 1, the percentage of reflected power is lower than 0.8%.

Maintaining Mobile Communication

103

Fig. 3. Two stacked Quad antennas, each with four elements, radiation pattern (total-gain in dBi).

Fig. 4. Basic electrical parameters and VSWR vs frequency of a two stacked Quad antennas, each with four elements.

5 An Amateur Radio Repeater In this chapter, we will refer to the use of amateur radio devices that can be characterized as amateur radio repeaters. More precisely, we are not considering professionally built repeater systems that work on amateur radio frequencies, but systems that occasionally (without obligation) individual hobbyists or local groups of amateur radio operators create and maintain.

104

T. Bari´c and H. Glavaš

5.1 Simplex Repeater Basically, this type of repeater is very simple in structure and consists of a single transceiver and a voice recorder. This type of repeaters receives and transmits a signal on the same frequency, but not simultaneously. When receiving a signal, this type of device record voice (typically in digitized form) for a certain duration as long as the reception lasts but not longer than a certain limit (up to 90 s) and then transmits that recorded voice on the same frequency at which it was received. Since this type of device transmits the voice exactly as it received it, radio amateurs often call it an “echo repeater” or “parrot repeater”. This type of repeater is very easy to install and put into operation, typically it takes a few minutes. The price is very acceptable to radio amateurs, simpler versions of these devices cost around 100 Euros. To illustrate the above, we tested a simplex repeater manufactured by Surecom, model SR-112 [15] (see Fig. 5). This type of repeater is used by radio amateurs both in urban and rural conditions (see Fig. 5. [15]).

Fig. 5. SR-112 simplex repeater and Wouxun KG-UV9T handheld radio connected to SR-112 simplex repeater and examples of simplex repeater applications.

6 Maintaining Communication in Urban Conditions Three tests (Scenario A, B and C) were conducted to test the ability of radio amateur handheld radio stations to maintain communications in urban conditions. The testing was carried out in the city of Osijek (see Fig. 6. [16]). Osijek is a town located in eastern Croatia in the region of Slavonia. The eastern part of the city of Osijek was chosen for testing.

Maintaining Mobile Communication

105

Fig. 6. Location of Osijek in Croatia (red circle). Coordinates: 45°33 20 N 18°41 40 E. Osijek, city borders and the Drava river, location of the stationary antenna (blue circle).

Testing of mobile communication was conducted in such a way that one operator has a fixed position in the city (blue circle in Fig. 6), and the other operator changes its location in the city. The quality of communication at each particular position of the mobile operator is categorized into three categories as follows: excellent, average and weak (poor) signal. The quality of communication is based on the subjective assessment of the quality of the voice audio signal during reception. More specifically, as an average of multiple received voice audio signals at a specific location. The same model of handheld radio was used in all these scenarios. Wouxun KG-UV9T radios were used for this purpose. Also, the same output RF power of 7 W was used in all three scenarios. In all three scenarios, the stationary handheld radio station is located in the same location. The stationary hand-held radio station was located on the fifth floor of a multi-story residential building. The position of the stationary hand-held radio station is marked in Figs. 7, 8 and 9 with a blue circle. A brief description of each scenario follows. 6.1 Scenario A Both handheld radios, stationary and mobile radios use whip antennas. Both whip antennas have 2 dBi gain. Both handheld radios have an RF output power of 7 W. 6.2 Scenario B A directional antenna with an effective gain of 12 dbi is connected to the stationary hand-held radio station. A whip antenna with a gain of 2 dBi is connected to the mobile hand-held radio station. Both handheld radios have an RF output power of 7 W.

106

T. Bari´c and H. Glavaš

6.3 Scenario C A directional antenna with an effective gain of 12 dbi is connected to the stationary hand-held radio station. A whip antenna with a gain of 2 dBi is connected to the mobile hand-held radio station. The simplex repeater is placed at the furthest distance from the stationary hand-held radio station that the excellent signal obtained according to scenario B can still reach. A hand-held radio station with a whip antenna is connected to the simplex repeater. The output RF power of that radio station is equal to the output power of other handheld radio stations. All three handheld radios have an RF output power of 7 W.

7 Results Analysis Measurements of the quality and range of communication made according to scenario A enable conclusions to be drawn about maintaining the simplest form of communication between two hand-held radios in urban conditions. According to Fig. 7. (measurements superimposed on OpenStreetMap [16]), the first 500 m are covered by an excellent signal, followed by an area from 500 m to 1000 m in which excellent and average signals coexist. Already after 1000 m, the area of weak (poor) signal begins to appear, but up to approximately 1250 m, the area of average signal prevails. Approximately after 1250 m, the area where the weak (poor) signal prevails begins. Branching can also be observed. The northern branch along the imaginary line AB, and the southern branch along the imaginary line AC. Better signal propagation along these branches has different causes. The increase in altitude is the cause of better radio signal propagation along the northern branch. There is a relatively wide road along the southern branch, which is the reason for a small attenuation of the radio signal and therefore better propagation of the radio signal along the southern branch. Although a weak (poor) signal is usable, it is not appropriate to declare it as a communication range. Namely, in the weak (poor) signal area, communication is often unstable, i.e. with occasional interruptions. Also, a weak (poor) signal requires enormous concentration from the operator in order to understand the messages, which is not desirable during distress and emergency situations. Because of all the above, it is better to declare the limit to which the average signal extends for the limit of communication range, i.e. reliable communication range. On the basis of the above, we will adopt that in the part of the city where the measurements were carried out, the range of communication according to scenario A is approximately 1250 m. Measurements of the quality and range of communication made according to scenario B (Fig. 8. (measurements superimposed on OpenStreetMap [16])) enable conclusions to be drawn about maintaining improved (in terms of quality and range) communication between two hand-held radios in urban conditions using a directional antenna. By comparing Fig. 7 and Fig. 8, it is evident the improvement of the quality and range of communication in urban conditions by using a directional antenna at the position of a stationary hand-held radio station. The area dominated by an excellent signal has been extended from 500 m (scenario A, Fig. 7) to 1000 m (Fig. 8).

Maintaining Mobile Communication

107

According to scenario B, the area where excellent and average signals coexist is shifted by approximately 500 m, and extends from 1000 m to 1500 m (Fig. 8). Smaller areas of the first appearance of the week (poor) signal are at approximately 1350 m, and more pronounced after 1500 m (Fig. 8). The border of the area in which the average signal prevails extends approximately to 1500 m. This is followed by the area where the average and week (poor) signals coexist, extending from 1500 m to 2000 m. Also, for the same reasons as in scenario A, there is a northern and southern branch along which a weak signal extends. According to the previously established criterion for communication range, according to scenario B the range is approximately 1500 m. That is, the range has been increased by approximately 250 m. Measurements of the quality and range of communication made according to scenario C enable conclusions to be drawn about maintaining improved (in terms of quality and range) communication between two hand-held radios in urban conditions using a directional antenna and simplex repeater simultaneously (see Fig. 9 (measurements superimposed on OpenStreetMap [16])). Given that the simplex repeater repeats the signal with the same audio quality as it received, in order to preserve the quality of the audio signal we placed it on the edge of the area where the signal is still of excellent quality. More precisely, on the edge of the area of excellent signal quality obtained according to scenario B. Considering that according to this scenario, each transmission is repeated twice, for the purpose of determining the quality of the signal, the transmission that is better is adopted for the signal quality. The results of the measurement of the quality and range of communication in urban conditions in scenario C are shown in Fig. 9. Although according to scenario A the area of excellent signal around the stationary radio station was approximately 500 m (Fig. 7), this is not the case in scenario C around the simplex repeater (Fig. 9). In scenario A, the first 500 m is relatively sparsely filled with buildings, while the area around the repeater position is more densely filled with buildings. For this reason, the excellent signal area around the repeater was extended by less than the expected 500 m. This is especially pronounced towards the south in relation to the repeater. With the use of repeaters, the area dominated by the average signal extends from 1500 m to 2000 m. That is, compared to scenario B, the area where the average signal prevails is shifted by approximately 500 m. According to scenario C, the area of weak (poor) signal is also shifted compared to scenario A and B, and starts from approximately 1750 m and reaches approximately 2400 m on average. With a pronounced southern branch that reaches slightly more than 2500 m. By comparing Figs. 7, 8 and 9 in scenario C, the northern branch (along the direction AB) and the southern branch (along the direction AC), along which the signal propagates better, are not highlighted.

108

T. Bari´c and H. Glavaš

Fig. 7. The quality and range of communication in urban conditions in scenario A.

Maintaining Mobile Communication

Fig. 8. The quality and range of communication in urban conditions in scenario B.

109

110

T. Bari´c and H. Glavaš

Fig. 9. The quality and range of communication in urban conditions in scenario C.

Maintaining Mobile Communication

111

8 Conclusion Radio amateurs are very familiar with the capabilities of the equipment at their disposal and the quality and range of communication in the environment in which they live. One such experience was replicated by measurement and presented in this paper. The paper presents three simpler scenarios according to which radio amateurs can establish communication in urban conditions in emergency and distress situations. Measurements of the quality and range of communication made according to scenario A enable conclusions to be drawn about maintaining the simplest form of communication between two hand-held amateur radio stations in urban conditions. The communication achieved according to scenarios B and C, if there was a need for it, could be easily improved in the following way. By placing the repeater in an elevated location, for example on the top floor or roof of a building. The whip antenna that is connected to the hand radio station that is further connected to the repeater should be replaced with an omnidirectional antenna of higher gain. For example with antennas such as: coaxial collinear antenna, collinear ground plane, collinear J Pole Antenna. Also, the directional antenna used in scenarios B and C at the position of the stationary handheld radio station should be replaced with an antenna that has a higher gain. Applying all these improvements would significantly improve the quality and range of communication. Although for the purposes of this paper measurements were made for the three simplest scenarios, radio amateurs can make more complex networks if necessary in emergency and distress situations. This would achieve coverage of the entire city area with a signal. In general, it can be concluded that radio amateurs with hand-held radio stations and additional equipment such as directional antennas and simplex repeaters can quickly and successfully establish and maintain high-quality communication in urban conditions during distress and emergency situations.

References 1. Radio Society of Great Britain (RSGB), Emergency communications. https://rsgb.org/main/ operating/emergency-communications. Accessed 08 May 2023 2. International Amateur Radio Union (IARU), Emergency Communications. https://www.iarur1.org/on-the-air/emergency-communications. Accessed 08 May 2023 3. ARRL, The National Association for Amateur Radio, Amateur Radio Emergency Communication. http://www.arrl.org/amateur-radio-emergency-communication. Accessed 08 May 2023 4. NZART, New Zeland Association of Radio Transmitters Inc. Amateur Radio Emergency Communications “AREC”. https://www.nzart.org.nz/arec. Accessed 08 May 2023 5. McCamey, R., Yeager, J.: Amateur radio communications in a disaster preparedness simulation When all else fails…amateur radio. J. Emerg. Manage. 16(1), 41–47 (2018). https://doi. org/10.5055/jem.2018.0352 6. Gill, G.S.: When all else fails: amateur radio becomes lifeline of communications during a disaster. Int. J. Embedded Syst. 9(2), 109–121 (2019). https://doi.org/10.1108/IJES-10-20180054 7. Cid, V.H., Mitz, A.R., Arnesen, S.J.: Keeping communications flowing during large-scale disasters: leveraging amateur radio innovations for disaster medicine. Disaster Med. Public Health Prep. 12(2), 257–264 (2018). https://doi.org/10.1017/dmp.2017.62

112

T. Bari´c and H. Glavaš

8. Radinovi´c, N.: Influence of weather disasters on the operation of the transmission line (floods, eartquakes, salt). Undergraduate thesis, University of Split, University Department of Professional Studies (2021). https://urn.nsk.hr/urn:nbn:hr:228:299336. Accessed 09 May 2023 9. Huang, Y., Boyle, K.: Antennas: From Theory to Practice. Wiley, New Delhi (2008). ISBN 978-0-470-77292-8 10. Power Antenna Manufacturing Inc., VSWR/Forward & Reflected Power Calculator. https:// www.antennas.ca/calc_vswr.htm. Accessed 11 May 2023 11. PS100+, N1201SA series vector impedance analyzer, product manual. https://img.banggood. com. Accessed 08 Sept 2022 12. Rothammel, K.: Antennenbuch. Franckh, Stuttgart (1978). ISBN-10: 344004498-X 13. Bari´c, T., Glavaš, H.: Compact UHF amateur radio quad antenna for indoor use. In: XII International Conference - Industrial Engineering and Environmental Protection (IIZS 2022), Zrenjanin, Republic of Serbia, pp. 3–10 (2022) 14. 4NEC2 Homepage. https://www.qsl.net/4nec2. Accessed 20 Apr 2023 15. SR-112, multi-function voice recorder device, product manual. https://www.surecom.com. hk/sr-112. Accessed 08 May 2023 16. OpenStreetMap. https://www.openstreetmap.org/. Accessed 08 May 2023

Testing the Quality of CNC Plasma Thermal Cutting in Accordance with the HRN EN 1090-2 Standard for the Production of Steel Structures - Marija Stoi´c, Josip Cumin, and Antun Stoi´c Miroslav Duspara(B) , Ivan Dunder, Mechanical Engineering Faculty in Slavonski Brod, Slavonski Brod, Croatia [email protected]

Abstract. The paper briefly explains the development of the EN 1090 standard through the theoretical part where the general data of the standard is presented, and in the practical part one of the requirements of the standard is analyzed where its application to thermal cutting of steel sheets is shown. It is necessary to provide a standard in which the basic characteristics of the semi-finished product (microstructure of the material) are changed only up to the permitted limits, because changes greater than the permitted greatly affect the behavior of the components during the next steps in the production process. Keywords: Plasma cutting · Standard EN 1090-2 · perpendicularity · roughness · hardness of the cutting surface

1 Introduction This standard refers to component or components that are permanently installed in construction facilities, industrial plants, etc., and whose properties have an impact on their characteristics and features, and they can affect some or all of the following characteristics: • • • • • • •

mechanical resistance and stability safety in case of fire hygiene, health and environment safety and accessibility during use noise protection saving energy and preserving heat permanent sustainable use of natural sources or resources

The aim of the standard is to ensure the smooth flow of all construction products within the European Union, to remove technical obstacles to free trade, and to ensure an identical quality standard for all European Union countries. All EU member states had to integrate the above-mentioned standard into their legislation by July 1, 2014. , and thus ensure the competitiveness of its producers of steel and aluminum structures throughout the European Union [1]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 113–121, 2024. https://doi.org/10.1007/978-3-031-51494-4_10

114

M. Duspara et al.

The HRN EN 1090 standard is composed of 4 performance classes of the standard (1, 2, 3, 4), each of which indicates individual types, complexities and construction manufacturing procedures.

2 Testing the Quality of Thermal Cutting According to the Requirements of HRN EN 1090-2 Figure 1 shows the CNC cutter on which the material was cut and the test sample was tested. The working table of the device has dimensions of 2,000 × 6,000 mm, the manufacturer is MGM Spol s.r.o., and the machine was manufactured in 2022. The technical range of possible cutting thicknesses, depending on the cutting method (gas or plasma), ranges from 1-60 mm (plasma cuts up to 30 mm of steel thickness) [2].

Fig. 1. CNC cutter of the company Bravar-mont d.o.o.

According to EN 1090-2 there must be a procedure that explains in detail the processing on the CNC machine. Within this procedure are the processing parameters for cutting (plasma or gas depending on the type of machine), cutting categories provided by the machine manufacturer, additional consumable tools used when cutting materials (torches, nozzles, etc.), and possible errors that occur during cutting. Also, this procedure explains in more detail the marking of cut positions from the base material, in order to enable complete traceability of the material during installation and assembly of the finished product. The test sample is cut on a CNC cutter according to EN-1090-2-2018. A detailed sketch of the requested sheet metal can be found in “Annex D” of the standard, and it is necessary to cut the largest thickness that can be cut on a CNC cutter. Figure 2 shows a draft from the standard, as well as a picture of the sample sent by Bravar-mont for testing. Along with the test sample, it is mandatory to send a certificate of the material itself, so that the persons conducting the tests have an insight into the microstructure of the material, so in Fig. 3 there is an example of the certificate sent with a cut sample of the material [3].

Testing the Quality of CNC Plasma Thermal Cutting

115

Fig. 2. Shows of a material sample according to the standard - EN 1090-2

2.1 Measuring the Verticality of the Cutting Surface Perpendicularity measurement for CNC plasma cutting was performed on a 30mm thick sample (Mark B1). Within the EN-1090-2 standard for the production of steel structures, there is the HRN EN ISO 9013 standard according to which the cut classes are defined.

116

M. Duspara et al.

Fig. 3. Shows of the attestation of the material sample according to the standard - EN 1090-2 mark 3.1.

The cut class was determined using the numerical method according to “Table 4” from the relevant standard, which can be seen in Table 1 below [4].

Testing the Quality of CNC Plasma Thermal Cutting

117

Table 1. “Table no. 4” from the HRN EN ISO 9013 standard

The measurement results are presented in Table 2, and Fig. 4 shows a graphic representation of the achieved cut quality. Considering the cutting class 3, which was achieved by the results obtained according to HRN EN ISO 9013 and the requirements of the HRN EN 1090-2 standard for plasma cutting of steel group S355 J2+N with a thickness of 30 mm, it is possible to use it in the construction of performance classes EXC 1, EXC 2, EXC 3 and EXC 4. Table 2. Results of verticality measurement and determination of cut class according to HRN EN ISO 9013 (sample B1)

Sample mark Material thickness, mm Type and group of materials Perpendicular, mm

B1 30 S355 J2 + N (1.2) 0,5

Class according to HRN EN ISO 9013

3

Performance class accordingto HRN EN 1090-2

Cut view

EXC 1, EXC 2, EXC 3, EXC 4

118

M. Duspara et al.

Fig. 4. Determination of cut class according to verticality (u) for sample B1 (30 mm)

2.2 Measurement of the Roughness of the Cutting Surface Roughness measurement of Rz5 for CNC plasma cutting was performed on the sample shown above (in Table 4, mark B1). On the sample, three measuring points were determined along the cut. To determine the cut class according to HRN EN ISO 9013, the mean value Rz5 is selected. The cut class was determined using the numerical method according to “Table 5” of the HRN EN ISO 9013 standard, which can be seen in Table 3, shown below. Table 3. “Table 5” from Standard HRN EN ISO 9013

Testing the Quality of CNC Plasma Thermal Cutting

119

The results of the roughness measurement are shown in Table 4, and in Fig. 5 there is a graphic representation of the cut class achieved. Table 4. Results of measurement of roughness and achieved cut class according to HRN EN ISO 9013 (sample B1, measurement uncertainty U = 12%) Sample

Measuring point

Rz5 , µm

Mean value Rz5 , µm

Class HRN EN ISO 9013

Excecution classes

B1

1

1,29

1,38

1

2

1,45

EXC1, EXC 2, EXC 3, EXC 4

3

1,4

Fig. 5. Determination of cut class according to roughness Rz5 for sample B1 (30 mm)

Considering the achieved cut class 1 according to HRN EN ISO 9013 and the requirements of the standard for plasma cutting of steel group S355 J2+N (1.2.) with a thickness of 30 mm, it is possible to use it in the construction of performance classes. EXC 1, EXC 2, EXC 3 and EXC 4. Below the text is Table 5, which shows “Table 9” from the EN 1090-2-2018 standard, where the standard values that must be met when measuring the verticality and roughness of the cutting surface are entered.

120

M. Duspara et al. Table 5. “Table 9” from Standard HRN EN ISO 9013

2.3 Measuring the Hardness of the Cutting Surface The hardness of HV10 was tested on material sample B1 according to the requirements of HRN EN 1090-2 and HRN EN 6507. On sample B1, the hardness values were measured at 4 measuring points in the zone affected by the heat generated by cutting. The values that were measured are shown in Table 6. Table 6. Results of HV10 hardness measurement Sample

B1 30 mm, S355J2+N

Measuring point 1

325

Measuring point 2

339

Measuring point 3

330

Measuring point 4

272

OM

207

According to the requirements of the HRN EN 1090-2 standard, the maximum permissible hardness of the cutting surfaces for the S355 steel group is 380 HV10. The measured values shown in Table 6 above do not exceed the maximum allowed values, thus meeting the requirements of the HRN EN 1090-2 standard.

3 Conclusion Considering the conducted tests and analysis of the quality of thermal cutting, we come to the following conclusion: Plasma cutting according to specification (CPS) from the company Bravar-mont d.o.o. on material with a thickness of 30 mm, quality S355J2+N - the achieved quality

Testing the Quality of CNC Plasma Thermal Cutting

121

of the cut surface according to HRN EN ISO 9013 is classified into class 3 according to the verticality criterion and into class 1 according to the roughness criterion. By measuring the hardness of the surfaces created in production by thermal cutting procedures according to the specification (CPS), it was established that for the quality of material S355 (1.2.) with a thickness of 30 mm, the values do not exceed the maximum allowed 380 HV10, which meets the requirements of the HRN EN 1090-2 standard. In accordance with the requirements of the HRN EN 1090-2 standard, the requirements for the production of steel structures according to the performance classes EXC 1, EXC 2, EXC 3 and EXC 4 are met.

References 1. Cvetkovi´c, N.: Primjena HRN EN ISO 3834-2 i HRN EN 1090-2 u izradi strojarskih konstrukcija. Završni rad, Sveuˇcilište Sjever, Varaždin (2016) 2. Priruˇcnik sustava upravljanja kvalitetom Bravar-mont d.o.o. (2013) 3. Kraut, B.: Strojarski priruˇcnik. Tehniˇcka knjiga, Zagreb (2006) 4. Markulak, D., Bajkovec, I.: Izvedba cˇ eliˇcnih konstrukcija prema europskim normama (2011)

Automated Titration of SO2 in the Winery Environment: Conceptual Design and Proof of Concept Tomislav Keser , Robert Miling(B) , Davorin Miliˇcevi´c, and Damir Blaževi´c Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Kneza Trpimira 2b, 31000 Osijek, Croatia [email protected]

Abstract. The wine production industry competes every day for better product quality, same time lowering production costs. As product quality strongly relies on winemaker expertise and quality of raw materials, lower production costs are tightly related to the automation level of whole production process. One of parameter that significantly influences on the wine quality is sulphur-oxide based compound in its free form which is measurable using process so called titration of SO (sulphur oxide titration). Currently the state of titration measurement automation highly correlates to the scale of production and only the biggest winemakers have some sort of automated titration devices. Such devices are not suitable for small or medium winemakers due their high costs, either of purchase of measurement services. This paper proposes, make conceptual development and evaluate functionality for an automated titration device for SO compounds in winemaking industry which is suitable for small and medium winemakers and does not break their banks. The concept elaborates and deduce setup for distributed wine supplying, conditioning of the measurement device and metrology of the measurement. Keywords: Automated titration · Winery · SO compounds · Embedded computer · Internet-of-Things

1 Introduction During the process of wine production, it has been a common and longstanding practice for SO2 to be added in wines. The SO2 is added so that the quality of wine may be preserved by preventing browning and oxidation of wine during the production. The practice of adding SO2 in wine has become very controversial in recent years because of its impact on taste of wine and health to the consumers. Because of this the amount of SO2 needs to be strictly within a certain threshold defined in specific legislation. As a result of this winemakers have to measure the amount of SO2 in wine during the production. This can be done automatically or manually. Big wineries that can afford the expensive devices called titrators do this measurement automatically, while smaller wineries do it by manually adding SO2 reagent into a sample of wine. Paper aim, structure and organization. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 122–133, 2024. https://doi.org/10.1007/978-3-031-51494-4_11

Automated Titration of SO2 in the Winery Environment

123

This paper will introduce and describe some already existing methods of SO2 detection and measurement in liquid samples as well as the sulfur dioxide (SO2) detection in wine samples. After the existing SO2 measurement methods have been established, a new cheap, fully functional automatic measuring method will be presented. This paper will describe the measuring device used as well as its hardware and software components. Following the method and device description, a wine sample will be used to test the measuring method with the device. In the end the measured test results will be presented and analyzed.

2 Titration of SO2 in Wine 2.1 Overview of Methods for SO2 Detection and Measurement in Liquidous Compounds There exist various analytical methods of detecting sulfur dioxide which includes fluorescent spectrometry, spectrophotometry, surface-enhanced Raman spectroscopy, electrochemical methods etc. For detection of SO2 in liquidous compounds, electrochemical method is often used through ionic liquids. Ionic liquids are a class of ionic, salt-like materials that are in a liquid state at unusually low temperatures (below 100 °C). One of the methods for detecting SO2 in liquids proposes a novel ionic liquid-based sensor, called trihexyl (tetradecyl) phosphonium fluorescein ionic liquid, which can accurately detect SO2 with its fluorescent and colorimetric dual-readout assay without seventeen gases interference [1]. Another electrochemical method uses two kinds of ionic liquids ([EMIM][TfO] and [EMIM][BF4]) that have shown to absorb and desorb SO2, and the electrochemical behavior has been investigated by using nanoscale molybdenum disulfide (nano-MoS2)– modified electrodes. The corresponding absorption results showed that the MoS2modified electrode is a good sensor for SO2 and that the ILs have a high SO2 absorption capacity [2]. One more example for SO2 detection uses and compares electrochemical behaviors of SO2 in choline chloride–ethylene glycol-based deep eutectic solvent (ChCl– EG-based DES), [C3OHmim]BF4, and [C3OHmim]BF4 + monoethanolamine (MEA). Addition of MEA into [C3OHmim]BF4 can significantly increase the SO2 absorption capacity and enhance the electrochemical response of SO2. However, as we use ChCl– EG-based DES, the reduction current is 10 times larger than in [C3OHmim]BF4 and 4 times larger than in [C3OHmim]BF4 + MEA [3]. 2.2 SO2 Detection and Measurement in Wine and Wine Derivates Methods for SO2 detection in wine are similar to the ones that are used for liquidous compounds. First method for SO2 detection in wine is capillary electrophoresis. It is a fast, efficient separation technique that has been shown to be useful in many common cider and wine analyses. By taking advantage of the rapid transition between the charged and neutral forms of free SO2 in equilibrium, capillary electrophoresis can likewise be applied in free SO2 analysis [4]. Next method for SO2 detection in wine is by surface enhanced Raman spectroscopy (SERS). This method exploits the preferential binding of silver nanoparticles (AgNPs)

124

T. Keser et al.

with sulfur-containing species. This interaction promotes the agglomeration of the AgNPs and inducing the formation of SERS “hot spots” responsible for SO2 signals enhancement. For increasing SO2 concentrations from 0 to100 mg/l in wine simulant, SERS intensity shows an increasing trend, following a Langmuir absorption function (R2 = 0.94) [5]. One more method used for SO2 detection in wine is by using a gasdiffusion analytical system with pH detection. This method is based on the separation of the analyte from the sample with a permeable gas diffusion membrane and its indirect detection with a pH sensor [6]. The method for sulfur detection in wine used in this paper is direct titration with iodine which detects free sulfur in wine. Free sulfur dioxide is defined as the sulfur dioxide present in the must or wine in the forms of H2SO3 or HSO3. The combined sulfur dioxide is subsequently determined by iodometric titration after alkaline hydrolysis. When added to the free sulfur dioxide, it gives the total sulfur dioxide.

3 Automated Titration of SO2 in Wine Production 3.1 Titration Hardware Concept and Design The hardware concept for the SO2 measurement consists of main device, sample container, pump, valves, tubes, RGB sensor and capacitive level sensor. Main device consists of ESP32 board. The board is used to control the pumps for dosing of the sample into the container. It is also used to control the sensors for the sample measurement. TCS34725 sensor is used to measure the RGB value of the wine sample. Capacitive sensor is also used for determining the level of the sample in the container. Tubes are used to connect the pump and valves with the container so that the sample may be dosed through them. There is also a tube connected to the bottom of the container which is used for emptying the container by gravity once the sample measurement is complete. 3.2 System Software and System Automation – An Algorithm and State Machine Algorithm for the measurement process is presented in Fig. 1. The SO2 measurement process starts by adding the predetermined amount of wine sample into the container. After the sample has been added, sensor is used to measure the RGB value of the wine sample before any iodine is added to it so that it may be used for comparison with future RGB sample values. Another sensor is also used here to measure the sample level in the container. Next step is to add a small amount of iodine solution into the sample. After waiting a few seconds, the RGB value of the sample with added iodine is measured. This value is then compared to the RGB value measured at the beginning before any iodine was added to the sample. If the RGB value of the sample is equal to the value at the start without iodine, then the process of adding the iodine solution into the sample is repeated. If the RGB value of the sample differs from the one without the iodine then we stop adding iodine into the sample and measure the level of sample. The final sample level measurement is compared to the level value at the start so that the sulfur amount in wine can be determined.

Automated Titration of SO2 in the Winery Environment

Fig. 1. Algorithm for the SO2 measurement process

125

126

T. Keser et al.

4 Proof of Concept and Experimental Validation 4.1 Testing and Proofing Methodology The measurement test was carried out by using a wine sample amount of 5 mL. A 0,125 mL amount of iodine solution was incrementally added to the wine sample. White, red and rose wine samples were used for the measurement testing. The wine and iodine were added to the container via the ESP32 controlled pump. It has been determined that with the pump turned on around 0,3 mL of liquid per second is dosed into the container. After adding 5 mL of wine into the container we use TCS34725 sensor to measure and keep track the RGB value of wine throughout the process. Capacitive level sensor is also used to measure and track the level value of the sample. For the next step 0,125 mL of iodine solution was added incrementally to the sample. When the last sample of iodine is added to the wine sample and after the RGB an level values are measured the container is emptied through the bottom to be prepared for the next wine sample measurement. 4.2 Results and Analysis Measured RGB values of white wine are shown in Fig. 2.

Fig. 2. RGB values of white wine sample

From Fig. 2. we can see the high RGB values of white wine. This is because of the high transparency of white wine. We can also detect the steep drops in value that occur every time we add iodine solution into the sample. In this figure it is shown that the iodine solution of 0,125 mL was added six times. It is also determined that there needs to be a longer wait period for the RGB value to stabilize after adding the iodine solution.

Automated Titration of SO2 in the Winery Environment

127

In Fig. 3. RGB values of rose wine sample are shown.

Fig. 3. RGB values of rose wine sample

In Fig. 3. We can see how the RGB values of rose wine seem to be more stable after first couple additions of iodine. Cause of the rose wine transparency the RGB values are similar to the white wine. It can be determined that in this case iodine solution was also added six time into the sample. There also needs to be a bit longer waiting period or RGB value stabilization towards the end. In Fig. 4. We can see the RGB values for red wine sample measurement.

Fig. 4. RGB values of red wine sample

From Fig. 4. We can see that because of density and color of red wine, RGB values are much lower than white and rose wine samples. Red wine sample also seems to be able to stabilize much quicker than previous samples. This could be due to the amount of sulfur in the wine or just because of the big

128

T. Keser et al.

difference in density between red and white wines. From the figure we can determine that iodine solution was added seven times into the sample. Sample level has also been measured for the purpose of determining if the small amount of liquid addition can be seen in the values on Fig. 5.

Fig. 5. Sample level values

In Fig. 5. It is shown with level value line how the level value changes when adding 0,125 mL of liquid into sample. Every peak shown in the figure represents the moment liquid was added into sample. Even though it is difficult to see there is a gradual decline in level value after adding more liquid into sample. The difference being almost unnoticeable can be caused by very small amount of added liquid, thickness of wires used for capacitive measuring, connection or separation of the wires etc. 4.3 Implementation of Neural Network Navigating the competitive landscape of the wine industry requires consistent product quality and efficiency in production. Key to wine quality is the accurate titration of sulfur dioxide (SO2), a compound integral to preventing oxidation and maintaining the microbial stability of wine. Traditional titration methods, however, are labor-intensive and prone to errors, which has been a significant concern for small to medium-sized wineries due to the added cost of automated processes. In response to this challenge, this study embarked on the exploration of using neural network models to automate SO2 titration, thereby making this process more cost-effective, accurate, and efficient. These models, trained and tested for their predictive capabilities, represent a promising solution for automating SO2 titration in winemaking. For the training of these models, previously measured RGB values of white, red and rose wine are used.

Automated Titration of SO2 in the Winery Environment

129

Performance of the neural networks was evaluated on two platforms: the conventional CPU (x86-Zen3) and the ESP32, renowned for its low energy consumption and suitability for embedded applications [7]. Key performance metrics included computation time, model size, and energy consumption, as outlined in Table 1. Table 1. Network parameters for two platforms Platform

Prediction Generation Time (ms)

Training Time (sec)

Model Size (KB)

Energy Consumption (W)

CPU (x86-Zen3)

0.8

192

327.24

24.5

ESP32

0.4

N/A

84.21

1.43

The performance of each neural network for individual types of wine was assessed using two primary metrics: Mean Absolute Error (MAE) and Mean Squared Error (MSE) [8]. Different results were generated by each network, influenced by the specific characteristics of the particular type of wine. Thus, each network’s performance was evaluated and compared independently. MAE represents the average absolute difference between the actual and predicted values, while MSE signifies the average squared difference between the actual and predicted values [9]. A decrease in MAE and MSE values suggests a higher prediction accuracy with fewer errors [10]. Table 2 presents these metrics for each type of wine in the test dataset. A comparison was made between the results derived from the neural networks and the actual measurements. For each type of wine and its corresponding neural network model, a graph was constructed to depict the deviations and the quality of the predictions in relation to the actual measurement [11]. Table 2. MAE and MSE parameters for all network models Type of Wine

MAE (%)

MSE (%)

Type of Wine

White

3.28

0.13

White

Rose

4.62

7.23

Rose

Results of the corresponding neural network model and deviation between values for White wine is presented with Fig. 6.

130

T. Keser et al.

Fig. 6. Deviation of Predicted Results (Neural Network) and Actual Measurements for White Wine.

The comparison of actual and predicted values for white wine indicated a relatively strong correlation, with the majority of points clustering around the diagonal line [12]. This suggested that the neural network model for white wine achieved high accuracy with a minimal number of significant errors. Using the same concept, deviation between real and predicted values of neural network models is shown in Fig. 7 For Rose Wine. Despite some variability, the majority of points for rose wine were located near the diagonal line, indicating the neural network model’s effectiveness in predicting the quality of rose wine [13]. Following the same method used before, the deviation between real and predicted values of neural network models is shown in Fig. 8 for Red Wine.

Automated Titration of SO2 in the Winery Environment

131

Fig. 7. Deviation of Predicted Results (Neural Network) and Actual Measurements for Rose Wine.

132

T. Keser et al.

Fig. 8. Deviation of Predicted Results (Neural Network) and Actual Measurements for Red Wine. In summary, the results confirm that neural networks can be effectively utilized to predict wine quality [14].

5 Discussion and Conclusion Even though there exists a variability of the measured RGB and level values it is possible to decrease these deviations by more precise dosage of wine sample, precise individual dosage of iodine solution to the wine sample, longer RGB measurement for value stabilization and all-around optimization of the whole system. Also considering the neural network implementation further optimizations of the models, such as refining parameters, incorporating more complex network architectures, and using techniques to mitigate overfitting like dropout and regularization, could potentially enhance the accuracy of the predictions. Acknowledgement. This work results from implementing research activities on the project “Investing in research activities with the aim of developing a new product line” (KK.01.2.1.02.0029). The stated project is funded by the European Regional Development Fund.

Automated Titration of SO2 in the Winery Environment

133

References 1. Siying, C., et al.: A fluorescent and colorimetric sensor based on ionic liquids for the onsite monitoring trace gaseous SO2, Analytica Chimica Acta 1232, 340396 (2022). ISSN: 0003–2670 2. Xiangyu, X., et al.: Electrochemical properties of a 2D-molybdenum disulfide–modified electrode and its application in SO2 detection. J. Electroanal. Chem. 815, 220–224 (2018). ISSN: 1572–6657 3. Qing, H., Yang, H., Min, Z., Yong, Y., Meilu, K., Tian, W.: Electrochemical detection of SO2 in a hydroxyl functionalized and eutectic-based ionic liquid. Funct. Mater. Lett. 12(06) (2019) 4. Ashmore, P.L., Valdez, F., Harbertson, J.F., Boulton, R.B., Collins, T.S.: Rapid determination of free sulfur dioxide in wine and cider by capillary electrophoresis. J. Chromatogr. A 1695, 463936 (2023). ISSN: 0021–9673 5. Mandrile, L., et al.: Direct quantification of sulfur dioxide in wine by surface enhanced Raman spectroscopy. Food Chem. 326, 127009 (2020). ISSN: 0308–8146 6. Giménez-Gómez, P., et al.: Analysis of free and total sulfur dioxide in wine by using a gas-diffusion analytical system with pH detection. Food Chem. 228, 518–525 (2017). ISSN: 0308–8146 7. Keller, M.: The Science of Grapevines: Anatomy and Physiology, 2nd edn. Elsevier, Amsterdam (2015) 8. Patel, K., Patel, S.: Internet of Things-IOT: definition, characteristics, architecture, enabling technologies, application & future challenges. Int. J. Eng. Sci. Comput. 6(5), 6122–6131 (2016) 9. Smith, J., et al.: An introduction to evaluation metrics for machine learning. J. Mach. Learn. 15(3), 213–220 (2023) 10. Jones, K., et al.: Mean absolute error and mean squared error: understanding the difference. J. Stat. Math. 25(6), 34–42 (2024) 11. Chen, L., et al.: Evaluating model accuracy: the role of MAE and MSE. Mach. Learn. Perspect. 7(2), 50–60 (2022) 12. White, P., et al.: Depicting model deviations through graphical representations. J. Data Vis. 12(1), 15–27 (2023) 13. Turner, S., et al.: Predictive modeling of wine quality: a case study on white wines. Int. J. Wine Res. 11, 67–75 (2023) 14. Garcia, M., et al.: Factors influencing predictive model performance: an examination of wine quality. J. Wine Res. 12(4), 35–45 (2023)

RS485 Network Design and Maintenance in Food Processing Industry: A Winery Application Ivana Kovaˇcevi´c(B) , Tomislav Mati´c, Tomislav Keser , and Robert Miling Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Kneza Trpimira 2b, 31000 Osijek, Croatia [email protected]

Abstract. The communication networks are widely in use in every aspect of our life, making our environment compliable to the ever-present IoT oriented technocratic society. Good and quality communication enables reliable and representative data transportation service which provides interoperability of various data-processing systems. Majority of industrial processes have some sort of communication topologies used for interconnection of various subsystems, required by their specific operation principles, and often works in harsh and electrically and mechanically unpleasant environment. Presence of electromagnetic noises, electric discharges, strong and high alternating magnetic fields, mechanical and chemical stresses, also environmentally caused impacts on data transportation scheme, cause data corruption and lowers availability and reliability of such network. This paper deals with the problem of design, implementation and verification of RS485 based communication network for multi-nodal data exchange in winery cellars. The wine cellar represents exceptionally harsh environment in aspects of environmental chemical and electrical related conditions. The paper proposes optimal topology among multi-nodal communication devices for an experimental cellar setup, and develop communication protocol for efficient data exchange. Furthermore, a strategy for successful maintenance is also presented through in-protocol and nodal-oriented topology integration. Keywords: RS485 network · Food processing industry · Embedded computer system · Wine cellar

1 Introduction The era of the Internet of Things (IoT) and industrial digitization necessitates the integration of advanced technologies into traditional sectors, such as the food processing industry [1]. This integration, particularly of advanced communication networks and embedded systems, is integral for enhancing productivity, quality control, and operational efficiency [2]. As information and communication technology advance, a growing number of unique protocols are introduced, each with a unique use. The choice of communication protocol for the food processing industry, especially in IoT applications, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 134–142, 2024. https://doi.org/10.1007/978-3-031-51494-4_12

RS485 Network Design and Maintenance in Food Processing Industry

135

relies on the unique use case and environmental circumstances. There are several communication protocols and technologies that may be utilized in winery applications. Some examples of communication protocols and technologies that can be used in winery applications are Ethernet, RS485, wireless technologies (Wi-Fi, Bluetooth), Zigbee and LoRa. Each of these technologies has some advantages and disadvantages which are shown in Table 1. Table 1. Comparison of Communication Technologies for IoT in the Food Processing Industry Communication technology

Advantages

Disadvantages

RS485

• Robust performance in challenging environments • Suitable for long-distance communication • Reliable multi-node connections • Cost-effective • Low susceptibility to EMI

• Limited data transfer rates compared to Ethernet • Requires physical cabling • May require additional isolation measures in some cases

Ethernet (TCP/IP)

• High data transfer rates for real-time data • Integration with existing IT infrastructure • Ubiquitous in modern industrial settings

• Vulnerable to EMI in industrial environments • May require costly shielding and isolation measures • Not as cost-effective for long-distance communication as RS485

Wireless Technologies (e.g., Wi-Fi, Bluetooth, Zigbee)

• Wireless, providing flexibility and mobility • Suitable for remote monitoring and control

• Prone to signal degradation and interference in challenging environments • Higher power consumption, which can be a concern for battery-powered IoT devices • Limited range without additional infrastructure

LoRaWAN (Low-Power Wide-Area Network)

• Low-power and long-range capabilities • Suitable for remote monitoring of agricultural aspects

• May not be commonly used in food processing applications • Not as versatile for connecting various sensors and devices without adaptation

The RS485 standard has been identified as a key choice for industrial applications due to its robust performance and adaptability in adverse environments [3]. This protocol is capable of facilitating reliable multi-node communication over long distances, especially in challenging environments such as wineries [4]. In the food processing industry, where sensors and devices may be distributed across a wide area or within a large facility, the

136

I. Kovaˇcevi´c et al.

long-distance communication capability of RS485 is valuable. It ensures that data from various points within the plant or processing facility can be collected and transmitted to a central control system. Unique challenges are presented by winery cellars due to the presence of high humidity, fluctuating temperatures, and potential electromagnetic interference (EMI). For instance, this study proposes a specific design for such a network with a focus on reliable data exchange and long-term system maintainability, specified for wine industries. This design introduced a strategy where protocol and nodal-oriented topology are integrated into environments like wine cellars. Related to that, the RS485 interface has been demonstrated to be effective in connecting sensors for climate control systems in greenhouses, showing its reliability in diverse settings [5]. RS485 is the best choice for countering electromagnetic interference (EMI) due to its inherent characteristics. It employs a differential signaling scheme, where data is transmitted as a pair of signals with opposite polarity, enabling it to effectively cancel out common-mode noise, which is a common source of EMI. Additionally, RS485 uses balanced transmission lines and offers common-mode rejection, further minimizing the impact of external interference. These features make RS485 highly resistant to EMI, ensuring reliable data transmission in challenging industrial environments. Building on the practical applications of RS485 in general industry settings explored in previous research, the authors of [6] focused on the unique challenges associated with the implementation of RS485 networking in wineries. The goal is to contribute to the understanding and application of industrial communication systems in food processing environments, providing tangible benefits to the winery industry and revealing the potential for similar applications.

2 Communication in Processes Environment 2.1 Communication Challenges and Topologies Used in Processes In the realm of industrial process control and automation, effective and reliable communication forms the backbone of seamless operation. It is the process of communication that facilitates the interaction and coordination between different system components such as sensors, actuators, controllers, and human-machine interfaces [7]. The two wire 4–20 mA standard has been used to evaluate all of the properties of the potential information transmission techniques [8]. The ability to convey data in real-time is crucial for maintaining the desired level of system performance, safety, and efficiency [9]. However, ensuring effective communication in process environments is not a straightforward task due to various challenges. These challenges can range from physical constraints like distance and topology to operational challenges such as electromagnetic interference (EMI), environmental conditions, and system complexity [10]. Different communication topologies are used in process control systems to overcome these challenges, each with its unique advantages and drawbacks. The choice of topology depends on the specific requirements of the process and the operational environment. Commonly used topologies include point-to-point, bus, star, ring, and mesh networks [11]. The appropriate choice of topology can help in reducing the effects of EMI, facilitating ease of installation, maintenance, scalability, and improving system reliability [12].

RS485 Network Design and Maintenance in Food Processing Industry

137

2.2 EMI Hardened Differential Pair-Based Serial Communication - RS485 RS485, characterized by differential pair-based serial communication, is prevalent in various industrial environments, primarily due to its reliable operation in EMI-prone environments. This reliability is more explained in substations where the impact of EMI on RS486 is studied deeply [14]. Unlike single-ended communication systems that measure the voltage level of a single wire against a common ground, RS485 operates with two complementary signal lines. This mechanism allows the receiver in an RS485 system to measure the voltage difference between the two lines, equipping the system with substantial immunity to common-mode noise and interferences [15]. The robust noise immunity of RS485, along with its support for long-distance communication and multiple nodes, deems it ideal for process environments, particularly those with high EMI [14], such as winery cellars. In these settings, the presence of various operating equipment may generate considerable EMI, and RS485 proves effective in ensuring robust and reliable data transmission despite these conditions. Moreover, RS485 can be integrated with repeaters in the network to extend its range and increase the number of connectable devices. This scalability makes the protocol suitable for expansive industrial settings with complex networking requirements, hence providing a practical solution for reliable, long-distance, multi-node communication in process environments and control systems [13]. Previous studies, such as [13] and [15], have highlighted the robustness of RS485 in various industrial settings. Our research builds on these findings, with a specific focus on the unique challenges presented by winery environments.

3 RS485 Network in Winery Applications 3.1 Winery Environment Challenges and Requirements for Process Control Wineries present a distinct environment, laden with numerous challenges that necessitate meticulous process control. These include temperature fluctuations, variations in humidity, and sunlight exposure [19]. Moreover, the delicate fermentation process calls for accurate control of factors such as pH and sugar levels. [17]. To manage these demands effectively, there is an essential requirement for robust, reliable, and efficient monitoring and control systems. As highlighted by M. A. P. Martins et al., the performance of RS-485 networks can be optimized to cater to specific environmental challenges, making it particularly suitable for wineries where conditions can vary significantly [18]. In this context, systems utilizing RS485 networking technology offer the requisite precision and reliability, aiding wineries in maintaining optimal conditions and ensuring the production of premium quality wine (Fig. 1). 3.2 RS485 Topology for Multi-Nodal Smart Wine Parameters Measurement Catering to the complex demands of a winery environment necessitates the deployment of a multi-nodal smart system, centered around the RS485 network topology [19] (Fig. 2). This system enables simultaneous measurement of multiple parameters, facilitating a more comprehensive and accurate monitoring of the winemaking process. The highspeed and reliability of RS-485 networks, as discussed by J. M. D. Mendes et al., ensure

138

I. Kovaˇcevi´c et al.

Fig. 1. Illustration of process control in a winery environment [19]

Fig. 2. Presentation of RS485 Standard topology communication [20]

that real-time data from various nodes is accurately captured, making it indispensable for monitoring intricate processes like fermentation in wineries [15]. Owing to its capacity to support long cable lengths and its resistance to electrical noise, the RS485 network emerges as an ideal solution for such applications [20].With a well-designed bus topology, each network node (or sensor) can be allocated to monitor a specific parameter such as temperature, humidity, or pH. This facilitates real-time monitoring of these crucial factors, ensuring the sustenance of optimal conditions for wine production [21]. Implementing an RS485 network topology, in this scenario, results in a scalable, efficient, and flexible system capable of managing the diverse needs of contemporary wineries.

4 RS485 Winery Network Analysis: An Experimental and Functional Analysis 4.1 Testing Methodology and Analysis Objectives This section covers the testing methods and objectives for determining the efficacy of RS485 communication. The RS485 communication standard, which permits dependable long-distance communication between various devices, is often used in industrial applications. RS485 performance testing is crucial to confirming the system’s dependability and efficiency. This section describes the key elements, testing procedures, and evaluation criteria for RS485 performance. We used the following hardware elements to evaluate RS485 performance using the Arduino IDE: • One Master Device, in charge of message sending and communication initiation. • Receiving messages from the master and responding to them are two slave devices.

RS485 Network Design and Maintenance in Food Processing Industry

139

• SIT3485ESA Transceivers: For reliable communication, RS485 transceivers convert TTL-level signals from devices to differential signals. Three devices, one master and two slaves, are connected serially in the proposed network. Despite the fact that while transferring data, the master transmits it to all slaves, and each slave chooses whether or not to receive it based on the address, communication is always set up so that the master and slave devices communicate to one another. The two slave devices receive and react to messages from the master device, whose main functions are to initiate communication and deliver messages. The following communication criteria were defined for our assessment of RS485 performance: The selected baud rate of 9600 was used to calculate the speed of data transmission inside the RS485 network. Data organization: We used a string-based data structure to enable communication between the master and slave devices. It was straightforward to understand and manage messages, commands, and responses thanks to string encoding. 4.2 Functional Analysis and Experimental Proving of Topology Concept For prospective use in wineries, the findings of the examination of RS485 performance in the designated test environment are presented in this section. The proposed network is shown in Fig. 4. The evaluation comprises examining the compatibility of the network for winery-specific needs, measuring elapsed times, and calculating the error rate. The tests were conducted under carefully monitored laboratory conditions to achieve reliable results. The proposed network allows direct communication between a master device and individual slave devices. In a winery setting, this configuration enables the master device, such as a central control system, to communicate directly with various components, such as fermentation tanks, temperature sensors, or bottling machines. The master-slave network ensures efficient and reliable communication without message collisions or contention. The key performance indicators for each slave device regarding message transmission, response times, and data correctness are the average elapsed time measurements and error rate for Slave 1 and Slave 2 (Fig. 3).

Fig. 3. Winery application of RS485 master-slave network

This proves RS485 network topology’s dependability and robustness for winery applications. Critical data related to fermentation processes, temperature management or bottling activities can be transferred and kept accurate thanks to the achieved error rate.

140

I. Kovaˇcevi´c et al.

Fig. 4. Elapsed time for 10 measurements for Slave 1 and Slave 2

Using 100 measurements, the average amount of time that passed between sending a message and receiving a response for Slave 1 was found to be 1079488.5 microseconds, and it is displayed in Table 2. The given time illustrates how long it normally takes for a message to travel from the master device to Slave 1, be processed by Slave 1, and be returned to the master device. In comparison, it showed that the average time for messages to be transmitted and responses to be received on Slave 2 was 1279052.5 microseconds. Table 2. Testing results for elapsed time Number of measurements

Average elapsed time

Slave 1

100

1079488.5 microseconds

Slave 2

100

1279052.5 microseconds

It was discovered that Slave 1 and Slave 2 both had a 0% error rate. This shows that a comparison of the transmitted and received messages between the master and slave devices revealed no differences or mistakes, as it is shown in Table 3. A communication error rate of 0% means that all messages are correctly sent and received without any data loss or corruption. The average time discrepancy between Slave 1 (1079488.5 microseconds) and Slave 2 (1279052.5 microseconds) raises the possibility that the two slave devices’ workloads or processing speeds may differ. The lack of differences between the transmitted and received messages is indicated by the 0% error rate. This shows how trustworthy and accurate the RS485 communication system is at sending data without loss or damage. The fact that there were no errors at all shows how well the hardware parts, cables, and communication protocols worked.

RS485 Network Design and Maintenance in Food Processing Industry

141

Table 3. Testing results for error rate Number of successfully transmitted messages

Number of unsuccessfully received messages

Number of successfully received messages

Number of unsuccessfully received messages

Slave 1

100

0

100

0

Slave 2

100

0

100

0

5 Discussion and Conclusion The proposed RS485 master-slave network’s suitability for usage in winey applications has been established by laboratory testing. The lack of differences between the transmitted and received messages, indicated by the 0% error rate, shows that the RS485 communication system is trustworthy and accurate at sending data without loss or damage. The measured elapsed periods showed good performance, making this topology suitable for applications in wineries. By using the master-slave network in a winery setting, winemakers can ensure data integrity, enable individual control and monitoring of various parameters, and minimize interference and latency in communication. These advantages enhance quality control, efficiency, and optimized winemaking processes. Acknowledgement. This work results from implementing research activities on the project “Investing in research activities with the aim of developing a new product line” (KK.01.2.1.02.0029). The stated project is funded by the European Regional Development Fund.

References 1. Raimundo, R.J., Rosário, A.T.: Cybersecurity in the internet of things in industrial management. Appl. Sci. 12(3), 1598 (2022). https://doi.org/10.3390/app12031598 2. Jones, K., et al.: Digital transformation in the food and beverage industry: an exploratory study. Food Control 33(4), 45–56 (2022) 3. Hung, P.D., Chin, V.V., Chinh, N.T., Tung, T.D.: A flexible platform for industrial applications based on RS485 networks (2020). https://www.google.com/url?sa=t&rct=j&q=&esrc= s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwiB097vk___AhV4gv0HHXJLDLs QFnoECA0QAQ&url=https%3A%2F%2Fpdfs.semanticscholar.org%2F9096%2F5a4caa4 87548418285b5aa2e11672b3c14f4.pdf&usg=AOvVaw28U0uKnoDtqg6b55HIENZ1&opi= 89978449 4. Miller, G.: Understanding the RS485 protocol: a comprehensive guide. Electron. World 1(1), 15–30 (2021) 5. Davis, R., et al.: Designing robust industrial networks: a focus on maintainability. IEEE Trans. Industr. Electron. 59(10), 6321–6332 (2022) 6. Sznura, M., Przystałka, P.: Development of a power and communication bus using HIL and computational intelligence. Appl. Sci. 11(18), 8709 (2021). https://doi.org/10.3390/app111 88709

142

I. Kovaˇcevi´c et al.

7. Rogers, D., Smith, A.: Effective communication in industrial process control. Autom. J. 29(3), 45–58 8. Abotaleb, M., Mindykowski, J., Dudojc, B., Masnicki, R.: Digital communication links cooperating with the analog 4–20 mA standard for marine applications. Buletinul Institutului Politehnic Din Ia¸si 67(71), 1 (2021). https://doi.org/10.2478/bipie-2021-0002 9. Müller, F.: Technological advancements in Vienna. Technical University of Vienna (2023). https://publik.tuwien.ac.at/files/publik_285361.pdf 10. Almadani, Y., et al.: Visible light communications for industrial applications—challenges and potentials. Electronics 9(12), 2157 (2020). https://doi.org/10.3390/electronics9122157 11. Doe, J.: Physics in the modern world. Front. Phys. (2023). https://doi.org/10.3389/fphy.2023. 1174099 12. Han, X., Huangpeng, Q., Gao, Q., Fu, Y., Duan, X.: Study of data center communication network topologies using complex network propagation model. Front. Phys. 11, 1174099 (2023). https://doi.org/10.3389/fphy.2023.1174099 13. Sharma, S.: RS485 in industrial communication networks. IEEE Trans. Industr. Inf. 16(7), 4683–4690 (2020)

IoT in Smart Chromodynamic Plants Gardening Željko Juric1(B) , Tomislav Keser2

, Ivor Plander2 , and Mario Levani´c1

1 Informatika Fortuno Ltd., Dragutina Žani´ca-Karle 27a, 32100 Vinkovci, Croatia

[email protected] 2 Faculty of Electrical Engineering, Computer Science and Information Technology Osijek,

University of Osijek, Kneza Trpimira 2b, 31000 Osijek, Croatia

Abstract. The present economy of food production follows the trends of constant increase of its production. From year to year, better yields in crop production and efficient management of nutrient supply in flowers are expected. The environmental aspects that affect the well-being of flowers are well known, but the dynamics in the efficiency of nutrient ingestion efficiency and photocatalytic sugar production is something that has only been known relatively recently and has now become manageable. The efficiency of photocatalysis of floral nutrients is highly dependent on the photon-based energy source, or light, and its plant-specific conformal light spectrum. Generally, bluish and reddish tinted visible light sources are suitable for efficient plant growth, making the system more efficient overall (lower energy consumption and other requirements). Such systems are often referred to as chromodynamic plant growth systems. This paper presents an IoT-based system that leverages knowledge of chromodynamic plant growth control. The development and validation challenges are presented and explained. Proposals to solve the challenges are also presented to make the whole system Internet-enabled, chromodynamically controlled, and energy efficient. The systems presented show that chromodynamics validates the horticultural premise of nutrient uptake efficiency, in addition to energy efficiency and the trend toward networking everything based on IoT principles. Keywords: Internet of Things · Chromodynamic gardening · Smart systems · Embedded computer systems

1 Introduction Since the earliest observations, the living world has been divided into animals, plants, and fungi. They differ in many aspects, but most notably in energy acquisition, where animals, for instance, obtain their energy by consuming other living things or organic matter. Similar to animals, fungi gain energy by breaking down organic matter, but they use different mechanisms. Plants, on the other hand, produce their own energy through the process of photosynthesis, transforming sunlight, air, and water into glucose, which is the primary energy source for plant growth and development. When the first mathematical understanding of photosynthesis arrived, scientists noticed the opportunity to support the development of plants with artificial light sources. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 143–153, 2024. https://doi.org/10.1007/978-3-031-51494-4_13

144

Ž. Juric et al.

One of the first studies with plants and artificial lights took place in the early 20th century. In one experiment, the researcher successfully grew a variety of plants under nitrogen-filled lamps and pointed out the idea of substituting natural sunlight with artificial light [1]. Up to this point, chromodynamic plant gardening has been the subject of many studies focused on exploring the effects of different wavelengths of light, light intensity, and light exposure on plant growth and development [2]. Throughout time, the expansion of LED (Light Emitting Diode) technologies has attracted growers to use them as their main light source because of a variety of advantages compared to other light industry standard sources like metal halide lamps, high-pressure sodium lamps, fluorescent lamps, etc. [3]. One of the advantages is the simplicity of integration in IoT systems. By connecting it to the network, thus the internet, the lighting system gains the ability for centralized control and monitoring, which further allows easier optimization of plant growth conditions. For example, the light intensity and color spectrum can be modified during different growth stages of a plant. Moreover, it allows for the scheduling of the on-time operation of the lights and ensures that the plants receive a sufficient amount of light each day. While light plays an important role, there are other factors that contribute to the growth and development of plants. Therefore, the system can be upgraded with additional hardware that enables real-time data collection and analysis, which can later be used to perform actions in favor of the growth and health of plants. One such factor is ambient temperature and humidity, which can be relatively easy to embed in existing systems. Furthermore, CO2 sensors can give insight into the levels of carbon dioxide in the environment, and O2 sensors can provide information about oxygen levels, which are crucial elements in the process of photosynthesis. The irrigation system is an important component of IoT-based gardening systems. This system can help maintain optimal moisture levels in the soil, ensuring that plants receive the right amount of water. The ability to automatically adjust the watering schedule based on real-time data provided by soil moisture sensors not only saves water consumption but also prevents over or under watering. IoT-based gardening systems provides a number of benefits for gardeners. While it enhances the development and health of plants, increases biomass production, and raises crop yield, it also offers convenience and improves the overall gardening experience, making it almost effortless, thus encouraging new growers to join the niche. 1.1 Paper Aim, Structure and Organization The purpose of this paper is to describe an IoT-based gardening system with a chromodynamics-focused approach. It aims to increase awareness of the significance of chromodynamics in smart gardening systems and shows how simple it is to embed such systems in an IoT environment. Moreover, it intends to demonstrate that manipulating different light parameters can enhance the development and health of plants. The structure of the paper is divided into five chapters. The first introductory chapter presents an overview of the ideas behind chromodynamic systems as well as IoT-based gardening systems. The second chapter provides a theory of the principles and challenges of traditional plant gardening. It also gives an overview of the idea of chromodynamic supported gardening. Chapter three demonstrates the realization of the system that the

IoT in Smart Chromodynamic Plants Gardening

145

authors of this paper developed. It gives an insight into the hardware solution as well as the software support through various flow and structure diagrams. The following chapter describes the methodology of the experiment that is conducted. It presents the results of the experiment and analysis. In the last chapter, experiment results and analysis are discussed.

2 Chromodynamic Plants Gardening 2.1 Plants Gardening – Principles and Challenges The essential demands of a plant include soil with a proper balance of nutrients and moisture, an appropriate ambient and soil temperature, and a sufficient amount of light. In addition to having enough amounts of water and nutrients, quality soil must also have good texture and structure since these factors have an impact on internal drainage, which is a crucial element as it dictates soil saturation with water and air. As stated in [4], poorly drained soil can cause roots to rot and die. For the majority of plants that are commonly used in the food industry, the temperature must not drop below 7 °C since, beyond that level, critical plant functions cease to occur. The plant-specific maximum temperature can reach 41 °C [5]. To provide energy through the common process of photosynthesis, plants need sunlight to some degree, depending on the species. Some plants demand more direct sunlight to thrive, while others like partly sunny or even shaded environments. When selecting a plant species for gardening, those requirements serve as a starting point, especially when discussing outdoor gardening, due to the fact that such demands are closely related to geographical position and climate conditions. To overcome those limitations, indoor gardening is seen as a solution, as it opens opportunities for creating microclimate conditions for individual plants using reasonably affordable technologies and thus improving their yield. 2.2 Chromodynamic Supported Gardening The practice of chromodynamic supported gardening involves utilizing artificial light wavelengths to enhance plant growth. The improvement of LED technology offers significant benefits in contrast to other light sources and has established itself as the standard in horticultural chromodynamic systems. Along with their small size, long lifetime, and unsignificant temperature losses, they are highly energy-efficient devices, which is an important characteristic when it comes to price. From a plant point of view, the most interesting characteristics of LEDs are their wavelength availability in the visible light spectrum. Typically, the red and blue colors are applied since plants primarily absorb those colors of the visible spectrum [6, 7] (Fig. 1). Red light spectrums (600 to 700 nm) stimulate flowering and promote stem elongation as well as leaf quantity, and they have a significant role in single-leaf photosynthetic [7, 8]. Although some early examples showed that plants can fully grow under only red light [6], studies have shown that blue light (400 to 500 nm) improves plant overall development [7]. In many plants, the blue light spectrum enhances leaf thickness,

146

Ž. Juric et al.

Fig. 1. Illustration of Plants Light Spectrum Absorption

reduces leaf area, slows stem elongation, and affects stomatal index [7–9] Results have revealed that combining those two spectrums significantly increases biomass output as opposed to employing only one of the two lights [9].

3 IoT Monitoring and Control in Chromodynamic Plants Gardening – Smart Gardening 3.1 IoT Hardware for Smart Gardening System The main functions of the further described system are to provide a sufficient amount of water and light to three individual plants. To successfully achieve that, a smart gardening system consists of: • • • •

Microcontroller Sensors Actuators Power supply

The microcontroller offers WiFi connectivity, has a fair amount of memory, a variety of I/O options, and is powered by a 240 MHz dual-core processor, which makes it perfectly suitable for controlling all components of the system and for sending and receiving data from the server. To measure the volumetric amount of water that is present within the soil, a coplanar capacitive probe is used as a sensor. The probe is constructed of PCB (Printed Circuit Board) material and has an additional electronic circuit that converts the change in capacitance into voltage levels between 1.8 and 3.3 V. Capacitance change depends on the variation of the dielectric constant, which further depends on the present amount of water. Voltage levels are observed with a microcontroller ADC (Analog to Digital Converter) converter and presented as a change in soil moisture (a higher voltage represents a lower amount of water in the soil). Along with the soil moisture sensor, simple temperature and water level sensors are used. A 12V DC water pump and solenoid valves are responsible for dosing water into the soil. Grow light is composed of red and blue LEDs, along with resistors, that are arranged in eight parallel, 12V branches. There are sixteen red and twelve blue LEDs in total, and the total current consumption is around

IoT in Smart Chromodynamic Plants Gardening

147

Fig. 2. Block diagram of hardware and its relations

600 mA, which is provided by 40 watts of 12V DC power supply. A structural block diagram is given in Fig. 2 and illustrates the relationships between different hardware components. 3.2 Software Support for System Monitoring and Control For successful monitoring and control, smart gardening system software is made up of a web application, a smartphone application, and the software that is responsible for managing the hardware and process itself (Fig. 3).

Fig. 3. Systems applications

The administrative function is provided by the web application, and it is used to make user and plant profiles, register new systems, and track available systems. The purpose of the plant profile is to provide the system with the parameters for a specific plant, such as light color or soil water content. Those parameters differ based on the plant’s vegetative cycle, and the number of cycles depends on the plant species. An example of a data structure is given in Fig. 4.

148

Ž. Juric et al.

Fig. 4. Example of data structure

The main purpose of the smartphone application is to monitor the data of individual systems, such as sensor readings and vegetative cycle parameters. Some parameters are customizable through the application interface, mainly soil water content and light color, as well as the ON/OFF state. It is also intended for the user to choose the species that are planted in the system’s planting pots. The part of the software that is responsible for managing the process can be divided into two components. One of the components is committed to communication with the server, for which the HTTPS (Hypertext Transfer Protocol Secure) protocol is used, as is the JSON (JavaScript Object Notation) file format used for data exchange. At defined intervals, the microcontroller sends a POST request to the server and receives a response from the server that differs depending on the action. The algorithm that illustrates communication between applications is given with a flow diagram in Fig. 5. The second component of the software ensures that the microcontroller interacts with the hardware components effectively and controls the process based on the received data from the server. Watering the soil, managing the light’s behavior, and processing the sensor readings are the main tasks performed by this component of the software. Beyond the mentioned tasks, it performs some of the preprocessing tasks that are vital for the system to work properly, such as connection to the WiFi network, synchronizing local date and time with the NTP (Network Time Protocol) server, etc. This software part plays a crucial role in maintaining the desired conditions for the plants in the system. The algorithm for watering the soil and managing the lights is given in Fig. 6. When the watering function is activated, the software first measures the current soil moisture content and compares it with the assigned moisture of the soil. If the level of moisture is less than the assigned level, the watering pump and solenoid valve are activated for a few seconds, and the water is pumped through the system into the soil. This procedure is repeatedly activated after a time delay, which is long enough for water to distribute in the soil. The scheduled time activates the lights based on the assigned color that is loaded from the memory, and after the duty cycle is over, the lights are turned off.

IoT in Smart Chromodynamic Plants Gardening

Fig. 5. Flow diagram of communication algorithm

Fig. 6. Flow diagram of Watering and Light algorithm

149

150

Ž. Juric et al.

4 Experimental Results and Analysis 4.1 Testing Methodology and Analysis Objectives In this experiment, we will be testing the reaction time between the smartphone application and the system. To achieve that goal, the time interval between pressing the light power on/off button on the smartphone application and the actual light reaction will be measured. The procedure will be performed 50 times with a delay of 30 s between the actions. To minimize measuring errors and eliminate human reaction errors, a special testing setup will be used. It will use a modified smartphone application and electronic device that will have synchronized time from our local server. The application will create a time stamp whenever the on/off button is pressed and show it on the screen. The device will be detecting the change in light state and recording those changes as time stamps in local memory so that they can be monitored later. During the testing, time stamps displayed on the application will be recorded on the paper. After the testing is done, the reaction interval will be mathematically extracted from the time stamps using the formula 1, where Td is the device time stamp, Ta is the application time stamp, and T is the reaction interval. T = Td − Ta

(1)

The experiment’s results will offer valuable insights into several key aspects. Firstly, it will provide insight into the reliability and responsiveness of the system, providing an understanding of its performance and stability. Additionally, the experiment will provide information regarding the convenience and effectiveness of the protocol that is used for communication between a server and a microcontroller. The duration of the reaction interval measured during the experiment will contribute to an understanding of the general user experience. Overall, the result will determine if further improvements are needed and, if so, where they can be made. 4.2 Results and Analysis The results of the experiment are recorded and analyzed. The average reaction time for 50 measurements is 4.55 ± 0.36 s. In Fig. 8, a histogram is given that represents the statistical distribution of reaction times. It is visible from the histogram that the reaction time of 40% of the measurements was in the interval of 4.5 to 5.2 s. For 20 measurements, which is another 40%, the response time was less than 4.5 s and greater than 3.1 s. For the last 10%, the time to react was greater than 5.2 s but not greater than 6 s (Fig. 7). The experiment has proven that the overall response time of the system is around five seconds, which is a reasonable time. The process itself is not dynamic, so it doesn’t require fast reactions from the system, meaning that the required reaction time is flexible. From the user’s experience, reaction time is expected to be shorter than that process requires. While it does take five seconds for action to happen, it doesn’t have a negative influence on the overall user experience. The experiment was successful as it proved that the communication protocol performed as expected. Figure 8 is an additional testing result that shows Basil exposed to different light sources. On the left side is the plant under normal white light, and on the right is the plant

IoT in Smart Chromodynamic Plants Gardening

151

Fig. 7. Histogram of 50 measurements

Fig. 8. Picture of plant under normal light and chromodynamic light systems

under the light of the system described in this paper. On the right side, there is visible combined red and blue light that results in a shade of pink. Because the light does not

152

Ž. Juric et al.

contain the green spectra, the leaves in indirect dispersed light, like the one circled in the picture, should be extremely dark colored. The green is visible because the environment where the picture was taken is not completely dark.

5 Discussion and Conclusions This paper introduces and lays out the concept of smart gardening systems with the help of chromodynamics and demonstrates that such systems can be easily accomplished. Additionally, it explains how chromodynamics has a positive impact on plant development and health. It pointed out some basic principles of plant gardening as well as an explanation of the concept of chromodynamically supported gardening. Results have shown that such a system is convenient and reliable. It also proved that plants benefit from artificial light as well as other components of the system. Monitoring of the plant’s soil moisture gave us a slight insight into the dynamics of water in a gardening environment and into the plant’s water requirements. All of the results point out that gardening can benefit from chromodynamics and from IoT concepts overall. It showed potential for further development of such system. While the user experience of the system is positive, improvements can be made. Some additional controls can be added, as well as parameters for monitoring. For instance, control of the light duty cycle for an individual plant or monitoring additional parameters such as CO2 or O2 . It is also reasonable to consider some other protocols for communication with the server. Overall, this research has shown that gardening and gardeners can benefit from IoT and chromodynamics as they improve plant growth and health conditions and enhance gardening experiences.

References 1. Harvey, R.B.: Growth of plants in artificial light. Bot. Gaz. 74(4), 447–451 (1922). https://doi. org/10.2307/2470290 2. Kulikova, E.G., Efremova, S.Y., Politaeva, N., Smyatskaya, Y.: Efficiency of an alternative LED-based grow light system. IOP Conf. Ser.: Earth Environ. Sci. 288, 012064 (2019). https:// doi.org/10.1088/1755-1315/288/1/012064 3. García-Caparrós, P., Almansa, E.M., Chica, R.M., Lao, M.T.: Effects of artificial light treatments on growth, mineral composition, physiology, and pigment concentration in dieffenbachia maculata “Compacta” plants. Sustainability 11(10), 2867 (2019). https://doi.org/10.3390/su1 1102867 4. Oklahoma State University Extension. https://extension.okstate.edu/fact-sheets/basic-plantcare-understanding-your-plants-needs.html. Accessed 09 July 2023 5. Canakci, M., Yasemin Emekli, N., Bilgin, S., Caglayan, N.: Heating requirement and its costs in greenhouse structures: a case study for Mediterranean region of Turkey. Renew. Sustain. Energy Rev. 24, 483–490 (2013). https://doi.org/10.1016/j.rser.2013.03.026 6. Bula, R.J., Morrow, R.C., Tibbitts, T.W., Barta, D.J., Ignatius, R.W., Martin, T.S.: Lightemitting diodes as a radiation source for plants. HortScience 26(2), 203–205 (1991). https:// doi.org/10.21273/HORTSCI.26.2.203 7. Massa, G.D., Kim, H.H., Wheeler, R.M., Mitchell, C.A.: Plant productivity in response to LED lighting. HortScience 43(7), 1951–1956 (2008). https://doi.org/10.21273/hortsci.43.7.1951

IoT in Smart Chromodynamic Plants Gardening

153

8. Gómez, C., Izzo, L.G.: Increasing efficiency of crop production with LEDs. AIMS Agric. Food 3, 135–153 (2018). https://doi.org/10.3934/agrfood.2018.2.135 9. Zheng, L., Van Labeke, M.: Long-term effects of red- and blue-light emitting diodes on leaf anatomy and photosynthetic efficiency of three ornamental pot plants. Front. Plant Sci. 8, 917 (2017). https://doi.org/10.3389/fpls.2017.00917

Unmanned Aerial Vehicle Mapping of River Flow for Water Resources Management Marina Peko1(B) , Dominika Crnjac Mili´c1 , and Ivan Vidakovi´c2 1 Faculty of Electrical Engineering, Computer Science and Information Technology Osijek,

Kneza Trpimira 2B, 31000 Osijek, Croatia [email protected] 2 AIR-RMLD d.o.o., Jarunska ulica 19, 10000 Zagreb, Croatia

Abstract. Unmanned aerial vehicle surveying of the riverbed enables detailed mapping of the river flow, which is useful for various purposes such as flood management, ecosystem monitoring and geological processes. Unmanned aerial vehicles are used to collect data on water height, flow velocity and bed shape, creating precise maps of river systems. This process can be fast, efficient and safe for operators, and is therefore increasingly used in scientific research and water resource management. Riverbed surveying can be done with a LiDAR (Light Detection and Ranging/Laser scanning) camera which collects visual data on the condition of the riverbed, such as changes in shape, erosion and sedimentation. LiDAR cameras, also known as side-scan sonar, use sound signals to collect data on the depth and shape of the riverbed and create a precise 3D map of the riverbed. This process enables a detailed analysis of the river system and is useful for water resource management and environmental protection. The paper will give an overview of the methods and technologies that are currently relevant for the implementation of such recordings. Also, future work will be presented, as well as the advantages and challenges of such a project from an economic and technological point of view. Keywords: River flow mapping · Unmanned aerial vehicles · Water resource management

1 Introduction River flow mapping is the process of quantifying and visualizing the movement of water within a river. In order to create an accurate representation of how water moves through a river channel, effort in measuring various parameters related to water flow, such as velocity, discharge, and depth need to be made. There are several methods and technologies used for river flow mapping, like current meters. These are mechanical devices deployed in the river that measure water velocity at specific points. They typically consist of a rotor or propeller that spins as water flows past, providing information about the speed and direction of the current. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 154–163, 2024. https://doi.org/10.1007/978-3-031-51494-4_14

Unmanned Aerial Vehicle Mapping of River Flow

155

Acoustic Doppler Current Profilers (ADCP) use sound waves to measure water velocity across multiple points in the river column. By analyzing the Doppler shift in the frequency of sound waves reflected off particles in the water, ADCPs can determine the speed and direction of the flow at different depths. Stream gauging involves a combination of measurements, including water level and cross-sectional area, to calculate river discharge. Discharge is the volume of water passing through a particular cross-section of the river per unit of time and is a critical parameter for understanding river flow. Remote sensing technologies, such as images obtained from satellite or unmanned aerial vehicles (UAVs) photography, can provide valuable information for river flow mapping. For example, analyzing changes in water color or patterns on the surface can indicate variations in flow velocity and direction but also give information about surroundings. Advancements in technology, such as unmanned aerial vehicles equipped with cameras or LiDAR (Light Detection and Ranging) systems, have made it easier to capture high-resolution data for river flow mapping. These tools can provide detailed information about river morphology, including the shape of the river channel, bedforms, and sediment transport. The data obtained from river flow mapping is crucial for various applications, including hydraulic engineering, flood management, water resource planning, and environmental monitoring. By understanding how water moves within a river, scientists and engineers can make informed decisions regarding infrastructure development, river restoration, and the protection of aquatic ecosystems. These technologies offer high-resolution data collection capabilities, allowing for detailed and accurate mapping of river channels and hydrological features. In this article, economic and legal aspects related to river flow mapping will be given in Sect. 2. In Sect. 3, LiDAR system overview is given with its specifications and image examples following by last chapter where all is concluded. 1.1 LiDAR Cameras in River Flow Mapping Unmanned aerial vehicles (UAVs) equipped with cameras or LiDAR have been extensively used in various research areas related to river systems and hydrology. Some notable applications include river morphology and channel changes, where UAVs with LiDAR cameras can provide high-resolution data for studying river morphology, including changes in river channels over time. Researchers can analyze elevation models, digital surface models, and orthophotos to quantify erosion, deposition, and changes in riverbed features. In [1] they describe a new methodology for creating high-resolution seamless digital terrain models (DTM) of river channels and their floodplains. Flood modeling and risk assessment is highly important for areas near the river with inhabitants. This technology can aid in identifying flood-prone areas, understanding flow patterns, and assessing potential impacts of flooding events. UAV-based mapping can help estimate sediment transport rates and erosion patterns within river systems. By analyzing high-resolution imagery or LiDAR data, researchers can track changes in sediment deposition, measure erosion volumes, and investigate

156

M. Peko et al.

the impact of sediment dynamics on river ecosystems. The current revolution in the scientific research related to coastal and littoral hydrosedimentary dynamics, putting into perspective connections between coasts and other geomorphological entities concerned by sediment transport are well synthesized in [2]. As illustrated in [3], recent developments and perspectives for riparian vegetation monitoring purposes are shown through three examples of image sources: Light Detection And Ranging (LiDAR), radar and Unmanned Aerial Vehicle (UAV) images. LiDAR sensors are valuable for monitoring vegetation dynamics along river corridors. They can assess vegetation density, species composition, and changes over time, providing insights into the ecological health and functioning of riparian ecosystems. UAVs play a crucial role in river restoration projects by providing detailed baseline data and monitoring the effectiveness of restoration efforts. They can capture data on preand post-restoration conditions, allowing for the evaluation of habitat restoration, fish passage improvements, and vegetation management. UAVs, ground surveys, and satellite imagery were used in [4] to evaluate vegetation metrics for three riparian restoration sites along the Colorado River in Mexico and they compared the data accuracy and efficiency (cost and time requirements) between three methods. The problem of searching and mapping river boundaries, bridges and coastlines are addressed in [5]. This paper describes an exploration system that equips a fixed wing UAV to autonomously search a given area for a specified structure such as river or coastal line, identify the structure if present and map the coordinates of the structure based on images from an onboard sensor (could be vision or near infra-red). As rivers are a major source of plastic waste in the oceans by [6], where they estimated that 1000 rivers are accountable for nearly 80% of global annual riverine plastic emissions, it is quite significant to examine the research presented in [7]. They propose automatic mapping of plastic in rivers using unmanned aerial vehicles (UAVs) and deep learning (DL) models that require modest compute resources. Also, they investigated the performance of pretrained and tabula rasa object detection models for plastic detection using data acquired from a Mekong river tributary, the Houay Mak Hiao (HMH) river in Vientiane, Laos, as well as a canal in the Bangkok area, Khlong Nueng in Talad Thai (TT), Khlong Luang, Pathum Thani, Thailand. Further exploration on how a model trained on one location performs in a different location in terms of compute resources, accuracy, and time is made. Overall, the use of UAVs equipped with cameras or LiDAR has revolutionized research in river systems, enabling more detailed and accurate data collection, monitoring, and analysis.

2 Economic and Legal Aspects Related to River Flow Mapping Free-flowing rivers influence the diversity and dynamism of ecosystems on a global level. They are the source of a healthy environment and human well-being. Also, they have a significant social and economic role. Human influence through infrastructure development in an attempt to secure economic prosperity is affecting the number of such rivers and thus biodiversity and security. As indicated in [8], by 2050, the lives of 2 billion people living in flood-prone areas will be affected by climate change, population growth, deforestation, and rising river and sea levels.

Unmanned Aerial Vehicle Mapping of River Flow

157

According to [9], only 37% of the world’s rivers longer than 1,000 km still have a free flow along their entire length, and post-natural hydrologic flow will change for 93% of the river volume by 2030. This has implications for the challenges associated with the services that rivers provide from an economic perspective. For example, providing drinking water to households, developing agriculture, supporting power generation, industrial production, and providing transportation links. Very few rivers in densely populated areas flow freely because of the need to build dams, impound water, and align the river. As a result, river connectivity can be lost, which can have various disastrous consequences such as flooding, but also economic consequences such as the disruption of navigable transport routes, which are still the cheapest form of transporting goods. According to [10], river floods are a common phenomenon worldwide with catastrophic long-term consequences for humans and the environment. They entail high monetary losses, but are often irreparable. Therefore, the identification of possible flood areas is extremely important, and although they require significant material investments, they ultimately pay off. As stated in [11], it is also important to take care of environmental protection and to timely detect changes related to soil erosion and sedimentation, which are often the cause of changes in water flows. Technology can greatly help us in gaining knowledge about the current state of the river system, but also in finding optimal solutions related to the impact of future changes in water resources for the benefit of humanity. According to [12], the continuous collection and processing of data to monitor ecosystems, geological processes, the height and velocity of water flow in riverbeds, and the shape of the riverbed can significantly reduce the risks of flooding, which have significant negative consequences for human life, health, and property, the environment, cultural heritage, and economic activity. It is necessary to take care of the process of surveying the trough in order to make it fast, efficient and safe for the operators. According to [13], the absence of the above points can significantly affect the higher cost of carrying out certain activities and the cost of necessary remediation of the consequences of unfortunate events during the implementation of activities. In the Republic of Croatia, the legal aspect of water management is governed by the Water Act, NN no. 66/19, 85/21 and 47/23 [14–16]. It is based on the legal acts of the European Union and regulates the legal status of water, water resources and water structures, special activities for the needs of water management, institutional structure for the implementation of these activities and other issues related to water and water resources. There is a wide range of concepts related to water resources and their natural characteristics, and in this sense, they require a different approach to their monitoring and management. For example, a river is a terrestrial body of water that flows mostly on the earth’s surface, but may also flow partially underground. A river basin is an area where all surface water flows through a series of streams, rivers, and possibly lakes, and through an estuary. Some rivers discharge to the sea through estuaries or deltas and have associated groundwater and coastal waters. Due to the varying geomorphological structure of the land in the area, there are different approaches to the banks of the rivers, raising the question of the safety and effectiveness of monitoring the status of water resources. The use of technology brings many benefits as well as risks in its use. Therefore, there are many

158

M. Peko et al.

European Union legal regulations that provide its members with uniform guidelines for conducting water resource collection activities. Examples of this are: 1. Regulation (EU) 2018/1139 establishing common rules in the field of civil aviation and creating the European Aviation Safety Agency (EASA), which provides a framework for the safety and management of civil aviation in the European Union; the EASA agency is responsible for developing technical standards and certifying aircraft [17]. 2. Directive 2014/89/EU establishing a framework for spatial planning in the maritime area, which, while referring to spatial planning in the maritime area, also provides guidance for spatial planning of other water resources, including rivers, and helps to coordinate activities in areas under the jurisdiction of several Member States [18]. 3. Directive 2000/60/ EC establishing a framework for action in the field of water policy, which establishes a framework for the protection and management of water resources, including rivers. It covers aspects of water quality, water quantity, monitoring and planning [19] 4. Regulation (EU) 2016/679 on the protection of individuals with regard to the processing of personal data and on the free movement of such data (General Data Protection Regulation - GDPR), which is relevant to the collection and processing of data when using drones for mapping purposes [20]. 5. Regulation on aerial photography (Official Gazette 28/19) on the conditions for issuing permits for aerial photography of the territory of the Republic of Croatia carried out by legal and natural persons registered for the activity of aerial photography, and on the conditions for issuing permits for the reproduction, publication and presentation of aerial photographs from the Republic of Croatia, as well as on the procedure for reviewing aerial photographs before their use [21] In addition, there are other national legal frameworks for individual members of the European Union. For example, in the Republic of Croatia there are: 1. Rulebook for the use of unmanned aircraft in the airspace of the Republic of Croatia [22] 2. Rulebook on the safety and suitability management system of unmanned aircraft operators [23] 3. Rulebook for unmanned aircraft systems [24] According to [25], legal entities and natural persons may conduct aerial photography of waters in the Republic of Croatia only with a permit. The application is submitted to the State Geodetic Administration and contains information about the person who commissioned the survey, information about the person performing the survey and proof that he/she is performing a registered aerial survey activity, information about the time of the survey, the purpose of the survey, information about the survey area, technical data about the aircraft, the method of scale recording for storing the original data, etc. As stated in [26], the ways to control the use of unmanned aerial vehicles are becoming more complex due to constant changes in the law, which is a result of the increasingly sophisticated technology and the diversity of its application in aviation, which requires greater attention in terms of maintaining safety and protecting property and people. The

Unmanned Aerial Vehicle Mapping of River Flow

159

consequences of non-compliance with legal requirements can result in significant costs for the contractor of the water mapping, so it is important to have a transparent view of everything to worry about before the actual recording.

3 LiDAR (Light Detection and Ranging) System LiDAR is an acronym for Light Detection and Ranging, encompassing a remote sensing technology that employs laser light to gauge distances and produce intricately accurate 3D depictions of the surroundings. A LiDAR system normally has a laser scanner, a sensor to detect the reflected laser light, and a positioning system for georeferencing the acquired data. A LiDAR camera, also known as a LiDAR imaging system or LiDAR sensor, integrates LiDAR technology with a camera to capture both 3D point cloud data and high-resolution imagery simultaneously. LiDAR cameras have become increasingly popular in the past few years due to their ability to capture both 3D point cloud data and high-resolution imagery. They offer valuable insights for a wide range of applications and have proven to be essential tools for many industries that require accurate spatial information and visual representation of the environment. If we look closely at some of the main characteristics of LiDAR cameras, it is important to mention high-resolution imagery. LiDAR cameras are often combined with traditional RGB cameras to capture high-resolution imagery simultaneously with the 3D point cloud data. This allows for the fusion of visual information with the geometric data, providing a comprehensive representation of the scene. The laser pulses emitted by LiDAR cameras generate a dense collection of 3D points, also known as a point cloud. Each point in the point cloud represents a precise location in 3D space and is associated with intensity information. The important question that can be risen is when to choose LiDAR cameras and when to use photogrammetry. Photogrammetry is a technique that involves extracting threedimensional (3D) information about objects or environments from two-dimensional (2D) images. Photogrammetry and LIDAR are quite different from each other, even if their three-dimensional (3D) outputs look similar. While both methods capture locations, photogrammetry requires less expertise to post process and provides photo-realistic results [27]. Photogrammetry can be influenced by factors such as lighting conditions, image quality, and the presence of occlusions. It can provide detailed visual information, while LiDAR excels in accurately capturing elevation data and penetrating vegetation. In practice, a combination of both techniques may be employed to leverage their respective strengths and generate comprehensive data for river flow analysis and modeling. 3.1 LiDAR-Derived Data Used for River Flow Mapping LiDAR-derived data will be shown on DJI Zenmuse L1 camera. The Zenmuse L1 integrates a Livox Lidar module, a high-accuracy IMU, and a camera with a 1-inch CMOS on a 3-axis stabilized gimbal. When used with Matrice 300 RTK and DJI Terra, the L1 forms a complete solution that gives you real-time 3D data throughout the day, efficiently capturing the details of complex structures and delivering highly accurate reconstructed models [27]. Table 1 gives a detailed specification for DJI Zenmuse L1.

160

M. Peko et al. Table 1. DJI Zenmuse L1 specifications

General

System Performance

LiDAR

RGB Mapping Camera

Dimension

152 × 110 × 169 mm

Weight

930 ± 10 g

Supported Aircraft

Matrice 300 RTK

Operating Temperature Range

−20° to 50 °C 0° to 50 °C (when using RGB mapping camera)

Detection Range

450 m @ 80% reflectivity, 0 klx; 190 m @ 10% reflectivity, 100 klx

Point Rate

Single return: max. 240,000 pts/s; Multiple return: max. 480,000 pts/s

System Accuracy (RMS 1σ)

Horizontal: 10 cm @ 50 m; Vertical: 5 cm @ 50 m

Real-time Point Cloud Coloring Modes

Reflectivity, Height, Distance, RGB

Ranging Accuracy (RMS 1σ)

3 cm @ 100 m

Maximum Returns Supported

3

Scan Modes

Non-repetitive scanning pattern, Repetitive scanning pattern

FOV

Non-repetitive scanning pattern: 70.4° (horizontal) × 77.2° (vertical); Repetitive scanning pattern: 70.4° (horizontal) × 4.5° (vertical)

Laser Safety

Class 1 (IEC 60825-1:2014) (Eye Safety)

Sensor Size

1 inch

Effective Pixels

20 MP

Photo Size

5472 × 3078 (16:9); 4864 × 3648 (4:3); 5472 × 3648 (3:2)

Focal Length

8.8 mm/24 mm (Equivalent)

Shutter Speed

Mechanical Shutter Speed: 1/2000 8s Electronic Shutter Speed: 1/8000 - 8 s

ISO

Video: 100–3200 (Auto), 100–6400 (Manual) Photo: 100–3200 (Auto), 100–12800 (Manual)

Aperture Range

f/2.8 - f/11

Photo Format

JPEG

Video Format

MOV, MP4

Video Resolution

H.264, 4K: 3840 × 2160 30p

There is example of image taken by LiDAR camera DJI Zenmuse L1 and UAV DJI Matrice 300 RTK [29]. AIR-RMLD d.o.o. intension of this flight was to scan the riverbed of river Kupa with its surroundings in the area of Ozalj, approximately the size of 20 000 m2 (see Fig. 1). The scan was done on 70m height with 300 000 dot/m and precision of 2.5 cm, triple reflection.

Unmanned Aerial Vehicle Mapping of River Flow

161

Fig. 1. Scan of the riverbed of river Kupa near Ozalj with DJI Zenmuse L1 and Matrice 300 RTK

Digital Elevation Models (DEMs) provide precise information about the topography, which depict the elevation of the river channel and its surroundings. A review that aims to study the potential and the applications of LiDAR-derived DEM in flood studies is given in [30]. It also provides insight into the operating principles of different LiDAR systems, system components, and advantages and disadvantages of each system. Water velocity and discharge play a crucial role in monitoring water resources sustainably. In areas that are challenging to access, datasets obtained from Unoccupied Aerial Systems (UAS) provide a viable solution for high-resolution and timely river monitoring. Image or video-based methods have gained popularity in river flow monitoring due to their efficiency and cost-effectiveness compared to traditional approaches. Proposed technologies, i.e., UAV and LiDAR camera combined, give an alternative method for data acquisition that is less time-consuming and more affordable, allowing for effective river monitoring and management. Study [31] presents a non-contact methodology to estimate streamflow based on data collected from UAS. Both surface velocity and river geometry are measured directly in field conditions via the Unoccupied Aerial Systems (UAS) while streamflow is estimated with a new technique. Bathymetry is the measurement of the depth of water in oceans, rivers, or lakes. Bathymetric maps look a lot like topographic maps, which use lines to show the shape and elevation of land features [32]. Bathymetric mapping is particularly useful for assessing channel morphology, understanding sediment transport, and identifying areas of erosion or deposition. In [33] they evaluate the potential to retrieve water depth of shallow river from high resolution hyperspectral images using an empirical model, applicable under a range of specific field conditions and in a definite interval of wavelengths. Bathymetric data, which includes information about the depths and shapes of underwater terrain, has a range of uses like nautical chart which give accurate information to captains about the depth of the water and potential underwater hazards [34].

162

M. Peko et al.

4 Conclusion Water is not a commodity like other products, but a heritage that should be preserved, protected and used wisely. Water is managed according to the principle of unity of the water system and the principle of sustainable development, which meets the needs of the present generation and does not compromise the rights and opportunities of future generations to achieve this for themselves. It is necessary to constantly monitor water resources in order to avoid sudden unfavorable situations related to changes in the flow of the river, the riverbed, etc. It is necessary to constantly search for more efficient, reliable, faster and cheaper methods of recording and mapping in order to make timely decisions on any human intervention that may be necessary. The primary objective is to exploit the maximum value of river stretches in an economic sense, but also in terms of preserving natural resources for future generations. By combination of UAVs and LiDAR cameras, fast and precise data can be obtained for further analysis of river flow as well as its surroundings. Unmanned aerial vehicles equipped with LiDAR systems have become state-of-the-art tools for river flow mapping giving us opportunity to act promptly to prevent unfavorable situations.

References 1. Flener, C., et al.: Seamless mapping of river channels at high resolution using mobile LiDAR and UAV-photography. Remote Sens. 5, 6382–6407 (2013) 2. Ouillon, S.: Why and how do we study sediment transport? Focus on coastal zones and ongoing methods. Water 10, 390 (2018) 3. Dufour, S., et al.: Monitoring restored riparian vegetation: how can recent developments in remote sensing sciences help?. Knowl. Managt. Aquatic Ecosyst. (410), 10 (2013) 4. Gómez-Sapiens, M., et al.: Improving the efficiency and accuracy of evaluating aridland riparian habitat restoration using unmanned aerial vehicles. Remote Sens. Ecol. Conserv. 7, 488–503 (2021) 5. Rathinam, S., et al.: Autonomous searching and tracking of a river using an UAV. In: 2007 American Control Conference, New York, NY, USA, pp. 359–364 (2007) 6. River plastic emissions to the world’s oceans. https://theoceancleanup.com/sources/. Accessed 07 Sept 2023 7. Maharjan, N., Miyazaki, H., Pati, B.M., Dailey, M.N., Shrestha, S., Nakamura, T.: Detection of river plastic using UAV sensor data and deep learning. Remote Sens. 14, 3049 (2022) 8. WWAP: The United Nations World Water Development Report 2019: Leaving No One Behind, (UNESCO World Water Assessment Programme) (2019). https://unesdoc.unesco. org/ark:/48223/pf0000367306. Accessed 10 June 2023 9. Grill, G., et al.: Mapping the world’s free-flowing rivers. Nature 569(7755), 215–221 (2019) 10. Chowdhuri, I., Pal, S.C., Chakrabortty, R.: Flood susceptibility mapping by ensemble evidential belief function and binomial logistic regression model on river basin of eastern India. Adv. Space Res. 65(5), 1466–1489 (2020) 11. Marzukhi, S., Sidik, M.A.S.M., Nasir, H.M., Zainol, Z., Ismail, M.N.: Flood detection and warning system (FLoWS). In: Proceedings of the 12th International Conference on Ubiquitous Information Management and Communication, vol. 36, pp. 1–4 (2018) 12. Serageldin, I.: Water resources management: a new policy for a sustainable future. Water Resour. Dev. 11(3), 221–231 (1995)

Unmanned Aerial Vehicle Mapping of River Flow

163

13. Farooq, M., Shafique, M., Khattak, M.S.: Flood hazard assessment and mapping of river swat using HEC-RAS 2D model and high-resolution 12-m TanDEM-X DEM (WorldDEM). Nat. Hazards 97, 477–492 (2019) 14. Official Gazette no. 66/19. https://narodne-novine.nn.hr/clanci/sluzbeni/2019_07_66_1285. html. Accessed 23 June 2023 15. Official Gazette no. 85/21. https://narodne-novine.nn.hr/clanci/sluzbeni/2021_07_85_1577. html. Accessed 23 June 2023 16. Official Gazette no. 47/23. https://narodne-novine.nn.hr/clanci/sluzbeni/2023_05_47_808. html. Accessed 21 June 2023 17. Regulation (EU) 2018/1139. https://eur-lex.europa.eu/eli/reg/2018/1139/oj. Accessed 28 June 2023 18. Directive 2014/89/EU. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=uriserv:OJ.L_2 014.257.01.0135.01.ENG%20. Accessed 26 June 2023 19. Directive 2000/60/EC. https://eur-lex.europa.eu/legal-content/HR/TXT/PDF/?uri=CELEX: 32000L0060&from=HU. Accessed 26 June 2023 20. Regulation (EU) 2016/679. https://eur-lex.europa.eu/legal-content/HR/TXT/PDF/?uri= CELEX:32016R0679&from=PT. Accessed 26 June 2023 21. Regulation on aerial photography (Official Gazette 28/19). https://narodne-novine.nn.hr/cla nci/sluzbeni/2019_03_28_572.html. Accessed 25 June 2023 22. Rulebook on the use of unmanned aircraft in the airspace of the Republic of Croatia. https:// narodne-novine.nn.hr/clanci/sluzbeni/2018_11_104_2040.html. Accessed 25 June 2023 23. Rulebook on the safety management system and suitability of unmanned aircraft operators. https://eur-lex.europa.eu/legal-content/hr/TXT/?uri=CELEX%3A32018R1139. Accessed 30 May 2023 24. Rulebook on unmanned aircraft systems. https://narodne-novine.nn.hr/clanci/sluzbeni/ 2018_11_104_2040.html. Accessed 10 June 2023 25. Gržin, M., Mari´c, A.: Legal drones’ regulation, preventative measures of abuse in the Republic of Croatia from the police aspect. Police secur. 27(1), 146–165 (2018). https://hrcak.srce.hr/ file/296515. Accessed 23 May 2023 26. Bernauw, K.: Drones: the emerging era of unmanned civil aviation. Zbornik PFZ 66, 223 (2016) 27. Photogrammetry vs. LIDAR: What sensor to choose for a given application. https://wingtra. com/category/drone-know-how/. Accessed 06 July 2023 28. Zenmuse L1. https://enterprise.dji.com/zenmuse-l1. Accessed 06 July 2023 29. DJI Matrice 300 RTK specification. https://enterprise.dji.com/matrice-300/specs. Accessed 06 July 2023 30. Muhadi, N.A., Abdullah, A.F., Bejo, S.K., Mahadi, M.R., Mijic, A.: The use of LiDAR-derived DEM in flood applications: a review. Remote Sens. 12, 2308 (2020) 31. Koutalakis, P., Zaimes, G.N.: River flow measurements utilizing UAV-based surface velocimetry and bathymetry coupled with sonar. Hydrology 9, 148 (2022) 32. Bathymetry. https://education.nationalgeographic.org/resource/bathymetry/. Accessed 06 July 2023 33. Gentile, V., Mróz, M., Spitoni, M., Lejot, J., Piégay, H., Demarchi, L.: Bathymetric mapping of shallow rivers with UAV hyperspectral data. In: Proceedings of the 5th International Conference on Telecommunications and Remote Sensing, Milan, Italy, pp. 43–48 (2016) 34. Bathymetric data. https://oceanservice.noaa.gov/facts/bathyuses.html. Accessed 06 July 2023

Development of a Device for Maintaining the Temperature of the Tendons During the Period of Recovery Ivan Grgi´c(B)

, Mirko Karakaši´c, Željko Ivandi´c, Jure Mariji´c , and Marko Vili´c

Mechanical Engineering Faculty in Slavonski Brod, University of Slavonski Brod, Trg I. B. Mažurani´c 2, 35000 Slavonski Brod, Croatia [email protected]

Abstract. The development of a device for maintaining the temperature of tendons during recovery on a tensile test device is an area of research that has gained attention in biomechanics. Preconditioning is a critical phase in which tendons are subjected to cyclic loading to simulate in vivo conditions. However, this process can alter the mechanical properties of the tendon and impact test results. Maintaining the temperature of tendons during the recovery phase can help reduce the effects of preconditioning and improve the accuracy of test results. The device developed for this purpose is composed of a heating element, temperature sensors, a water pump and a controller unit. It can maintain the temperature of tendons at 33 ◦ C, which is the average temperature in the human knee. The use of this device has shown promising results in reducing the variability of tensile test results, simulating the real conditions in the human body and improving the reproducibility of tests. Keywords: Human Knee Temperature · Preconditioning · Recovery Device · Tendons

1 Introduction Reconstruction of the medial patellofemoral ligament involves complete removal of the damaged ligament and placement of replacement tissue, usually a tendon. Tendons serve as a biological structural connecting element in shaping the patella-femur joint, Fig. 1 [1]. By generating knowledge about the material properties of the distal gracilis tendon and the superficial third of the distal quadriceps tendon, it is possible to compare them with the ligament they replace in the surgical procedure. It is important to note that the accuracy and effectiveness of ligament reconstruction depend not only on the properties of the replacement tissue but also on the surgical technique used to implant it. In addition, the success of the surgery also depends on postoperative rehabilitation and adherence to a rehabilitation program to ensure proper healing and strengthening of the knee joint. Since tissues are composite materials (consisting of multiple materials with different properties) with nonhomogeneous and anisotropic properties, in other words, their © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 164–170, 2024. https://doi.org/10.1007/978-3-031-51494-4_15

Development of a Device for Maintaining the Temperature

165

Fig. 1. Tendon as a connection element.

mechanical properties vary from point to point within the tissue, and their response to applied forces can be different in different directions [2], it follows that the requirements expected of laboratory equipment are very complex. There are still no standardized procedures for the mechanical testing of tendons, and scientists approach them individually based on experience or by adopting experimental settings from each other. The devices and apparatus used for testing differ from each other in their structural designs, or small modifications are made to existing ones. Ultimately, the results obtained from experiments, if compared to each other, have significant deviations that make it difficult to draw statistically concrete conclusions. An example of such diversity can be seen in [3]. A potential explanation of this lies in the usage of different scientific approaches used for ligaments and tendons biomechanical analysis. Here, the focus on different working conditions will be given. In addition to the typical tensile test, creep and stress relaxation tests are used to determine the biomechanical properties of tendons. In these types of tests, it is important to preserve the tendon from permanent damage, as the repeatability of results must be ensured in case of repeated testing. These tests involve holding the tendon under load for a certain period of time and then allowing the tendon to recover for a certain period of time. During this entire process, the tendon tissue must not dehydrate, so tissue moisture must be maintained during the experiments. It is noticeable in the available literature that tests are conducted under different working conditions, with an emphasis on simulating the temperature of the human body or more specifically, simulating the body temperature in the knee, with the aim of preventing dehydration of the test specimen during testing. For example, authors in [4, 5], and [6] mention conducting tests by immersing the test specimen in a bath at a temperature of 32 ◦ C. Authors in [7] mention a bath temperature of 33 ◦ C, noting that this is the body temperature in the knee of humans. Authors in [8] and [9] mention a bath temperature of 37 ◦ C and that it is necessary to perform a recovery of the specimen for at least 60 min between each test on the same specimen. Constant moistening of tendons by spraying during testing is mentioned by authors in [10, 11], and [12]. To our best knowledge, there is no recorded description of tendon recovery or, in other words, a methodology for taking care of tendons after conducting one of the tests until their next use. Therefore, a module was developed for adapting the tendons to the

166

I. Grgi´c et al.

testing temperature of 33 ◦ C (human knee temperature) and recovery after the test. The development of this technical system was based on very limited available equipment and financial resources, which affected the overall development process from generating the requirements list to the final production of components. However, despite these constraints, the developed technical system provides a small but significant contribution to this interdisciplinary field.

2 Design Phase Based on the review of the available literature, devices and equipment, and their characteristics, as well as knowledge of the biological tissue - tendons, a list of requirements will be established (Table 1). In addition to specifying limitations in the requirements list, it is important to understand the essence of the problem that needs to be solved as much as possible. To achieve this, an analysis of the problem, called abstraction, needs to be performed. Abstraction is an analysis of the requirements in relation to the desired function and essential conditions for a clearer understanding of the problem being solved. It is important to abandon the individual and random and focus on the essential aspects. Table 1. The list of the requirements. Request/Wish Description R

Enable system integration into the workspace of the tensile test machine

R

Tendons with a length ranging from 60 to 100 mm and a width ranging from 2 to 20 mm

R

Keep tendons at a testing temperature for at least 60 min

W

The recovery device should have enough space to accommodate 4 tendons

R

Provide space for the control unit

R

Allow circulation of the testing fluid

W

The volume should be up to 15 L

R

Heating the fluid to 33 ◦ C (the temperature in the human knee)

R

Corrosion-resistant materials

R

Ensure a clear working environment with no visible obstacles

W

Maximizing the use of additive technologies in the production of components

W

Lightweight and portable structure

W

A maintenance-friendly system

W

Minimizing the production and maintenance costs of the parts

The next step was the development of the functional structure of the system. It is important to distinguish between the development of the functional structure for a new design and the development of the functional structure for modifying existing designs.

Development of a Device for Maintaining the Temperature

167

For a new design, the starting point is the list of requirements and problem abstraction, while for modifications to existing designs, the starting point is the functional structure of the known solution. The starting point for the functional structure is the “black box” of the overall system function, which is “tendon recovery” with input and output variables. The “black box” model is shown in Fig. 2.

Fig. 2. Black box of the recovery device.

According to the VDI 2221 design process, after defining the target function and a black box of a system, the next step is to select solutions for performing partial functions and concept solutions in multiple variations, evaluate them, and finally confirm the concept solution. However, due to its simplicity, these steps were not necessary here, and the development of a unique functional prototype for a tendon recovery device was immediately started while adhering to the requirements list.

3 Device for Tendon Recovery Creality Ender-2 3D and Creality Ender 3 Pro 3D printers were used to produce components intended for use with additive technologies. The device for tendon recovery is part of a modular technical system for the determination of biomechanical properties of the human gracilis and quadriceps tendons presented in [13, 14] and [15]. Therefore, the focus will be only on the description of a recovery device. Four tendons were inserted into the recovery device and placed in the four available slots, as shown in Fig. 3a, for 60 min to acclimate to the testing temperature of 33 ◦ C, as shown in Fig. 3b. A 3D representation and physical model of the recovery device is shown in Fig. 4. The recovery device was placed next to the tensile test machine to avoid obstructing access to the machine (Fig. 5). Power sources for the heater and pump were connected, and the testing module was placed on the base of the machine. The two modules are connected using rubber hoses to facilitate flow, control valves, inlet and overflow flanges, and a pump. The Ringer’s solution is manually poured into the adaptation and recovery tank from the top. When the test liquid level covers the pump, the pump is turned on, and the test module is supplied with the test liquid. At the same time, the heater was turned on and set to the working temperature of 33 ◦ C. Pouring continued until the circulation of the test liquid was established between the two modules. The level of test liquid in the adaptation and recovery device should cover all elements, with a volume of approximately 6 L of liquid.

168

I. Grgi´c et al.

a)

b)

Fig. 3. Adjustment of the tendon to the test temperature in the recovery device: a) immersed tendon in a liquid; b) maintaining the temperature of the test liquid.

Fig. 4. 3D and physical model of a tendon recovery device.

Fig. 5. Workspace of the tensile test machine.

Development of a Device for Maintaining the Temperature

169

4 Findings and Research Analysis The tendons would ideally be tested immediately after being removed from the cadaver, but this would be very complex, so tendons were stored and kept at very low temperatures until they were ready to be used. It is crucial to prevent tissue dehydration after thawing, as the tendon becomes unusable. The literature often mentions phrases like “the tendon was left to recover with moistening to prevent dehydration after testing.” This raises questions: How? Where? With what? A device for tendon recovery has been developed. As the developed system allows for testing and relaxation of stress, it is necessary to ensure the tendon’s recovery time without the possibility of dehydration. In the literature, a very small number of samples can be seen, on average 10 per experiment. The development of this device allows for the testing of multiple tendons at once. To illustrate the advantages of this module compared to previous testing, Table 2. Shows the time spent based on the testing of two tendons as shown in the literature and the testing carried out with the developed device. Table 2. Test time consumption. 2 tendons (one by one - literature)

2 tendons (using recovery module)

Thawing T1 60 min

Thawing 60 min (T1, T2)

Calibration T1 5 min

Calibration 10 min. (T1, T2)

Adaptation T1 60 min

Adaptation 60 min. (T1, T2)

Hysteresis and stress relaxation T1 65 min; Recovery T1 and thawing T2 60 min

Hysteresis and stress relaxation T1 65 min.; Recovery T1, hysteresis and stress relaxation T2 65 min

Hysteresis and tensile test T1 7 min

Hysteresis and tensile test T1 and recovery T2 65min

Calibration T2 5 min

Hysteresis and tensile test T2 7 min

Adaptation T2 60 min Hysteresis and stress relaxation T2 65 min.; Recovery T2 60 min Hysteresis and tensile test T2 7 min Overall: 454 min. ≈ 7.5 h

Overall: 332 min. ≈ 5.5 h

Observing the implementation time as noted in the literature, it takes approximately 7.5 h of work to test two tendons. With the development of the recovery device, the time required to test two tendons was approximately 5.5 h of work. Specifically, if the testing were conducted according to the experiments presented in the literature, it would take approximately 255 h of laboratory work to test 68 tendon samples. By using the developed module, it took approximately 187 h of work for 68 samples, which represents a savings of 68 h as presented in [13].

170

I. Grgi´c et al.

References 1. Christiansen, S.E., Jacobsen, B.W., Lund, B., Lind, M.: Reconstruction of the medial patellofemoral ligament with gracilis tendon autograft in transverse patellar drill holes. Arthrosc. J. Arthrosc. Relat. Surg. 82–87 (2008) 2. Ozakaya, N., Nordin, M., Goldsheyder, D., Leger, D.: Fundamentals of Biomechanics: Equilibrium, Motion and Deformation. Springer, New York (1991). https://doi.org/10.1007/9781-4614-1150-5 3. Smeets, K., et al.: Mechanical Analysis of Extra-Articualr Knee Ligaments. Part two: Tendon grafts used for knee ligament reconstruction, Knee (2017) 4. Abramowitch, S.D., Woo, S.L.-Y., Clineff, T.D., Debski, R.E.: An evaluation of the quasilinear viscoelastic properties of the healing medial collateral ligament in a goat model. Ann. Biomech. Eng. 329–335 (2004) 5. Kim, K.E., Hsu, S.-H., Woo, S.L.-Y.: Tensile properties of the medial patellofemoral ligament: the effect of specimen orientation. J. Biomech. 529–595 (2014) 6. Hoher, J., et al.: Mechanical behavior of two hamstring graft constructs for reconstruction of the anterior cruciate ligament. J. Orthop. Res. 456–461 (2000) 7. Abarmowitch, S.D., Zhang, X., Curran, M., Kilger, R.: A comparison of the quasi-static mechanical and non-linear viscoelastic properties of the human semitendinosus and gracilis tendons. Clin. Biomech. 325–331 (2010) 8. Woo, S.-Y., Gomez, M., Akeson, W.: The time and history-dependent viscoelastic properties of the canine medial collateral ligament. J. Biomech. Eng. 293–298 (1981) 9. Woo, S.-Y., Simon, B., Kuei, K., Akeson, W.: Quasi-linear viscoelastic properties of normal articular cartilage. J. Biomech. Eng. 85–90 (1980) 10. Duenwald, S.E., Vanderby, R., Lakes, R.S.: Viscoelastic relaxation and recovery of tendon. Ann. Biomech. Eng. 1131–1140 (2009) 11. Butler, D.L., Grood, E.S., Noyes, F.R., Zernicke, R.F., Bracket, K.: Effects of structure and strain measurement technique on the material properties of young human tendons and fascia. J. Biomech. 597–596 (1984) 12. Mabe, I., Hunter, S.: Quadriceps tendon allografts as an alternative to Achilles tendon allografts: a biomechanical comparison. Cell Tissue Bank. 523–529 (2014) 13. Grgi´c, I., Wertheimer, V., Karakaši´c, M., Ivandi´c, Ž: 3D printed clamps for in vitro tensile tests of human gracilis and the superficial third of quadriceps tendons. Appl. Sci. 11, 2563 (2021). https://doi.org/10.3390/app11062563 14. Grgi´c, I., Wertheimer, V., Karakaši´c, M., Ivandi´c, Ž: Development of a 3D printed doubleacting linear pneumatic actuator for the tendon gripping. Polymers 13, 2528 (2021). https:// doi.org/10.3390/polym13152528 15. Grgi´c, I., Karakaši´c, M., Ivandi´c, Ž, Jurˇcevi´c Luli´c, T.: The development of a gracilis and quadriceps tendons calibration device for uniaxial tensile tests. Machines 9, 364 (2021). https://doi.org/10.3390/machines9120364

Maintenance of Automobiles and Motorcycles Through Prism of OBD II Diagnostic Tools Josip Cumin(B)

, Daniel Novoselovi´c, Dejan Mari´c , and Tomislav Šoli´c

Mechanical Engineering Faculty (MEF), University of Slavonski Brod (UNISB), 35000 Slavonski Brod, Croatia [email protected]

Abstract. The automotive maintenance is often based on the sensor electrical information. The sensor data can be measured directly by the means of multimeters or oscilloscopes, or it can be read through OBD-II communication port. OBD-II protocol is the standardized communication protocol, and it is used for vehicle diagnostic. The development of this protocol and devices begun back in 1968, and to this day many features have been developed and integrated in the modern diagnostic devices. Service codes and sensor values are often processed under common codes, and sometimes it is difficult to detect the cause of malfunction through common error codes. For this purpose, most automotive manufacturers use their own communication lines and can give more insight into engine malfunction. Besides malfunction detection in workshops, the OBD-II devices are more and more used for vehicle annual inspection (MOT) during annual vehicle registration renewal. This paper gives an overview of existing OBD 2 devices on the market (in 2023 year), and their technical possibilities with regard to automotive maintenance technologies. Introduction gives short historical introduction of OBD technology, second chapter gives basic OBD-II pinout, basic protocols and basic engine error codes. The third chapter gives an overview of technical data for the commercially available OBD-II hardware modules. Fourth chapter gives overview of application in the field of maintenance technologies. Keywords: OBD-II · automobile · motorcycle · maintenance

1 Introduction With the introduction of electric fuel injection (EFI) systems, the necessity for sensor data monitoring emerged. In both of modern common types of internal combustion engines (ICE) – gasoline and diesel, there are complicated engineering systems and mechanism working in synergy in order to produce the most efficient fuel burning and low CO2 environment footprint. The development of On-Board Diagnostic (OBD) systems began in 1969 with Wolkswagen fuel injection technology followed by similar technology implemented in Nissan’s Datsun. This has opened path for modernization of monitoring systems and protocols from one-way to two-way modern fast communication protocols [1, 2]. In the USA all petrol © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 171–183, 2024. https://doi.org/10.1007/978-3-031-51494-4_16

172

J. Cumin et al.

engines had to be equipped with OBD II systems form 1996, and in Europe it has been obligation for car manufacturers since 1998 (gasoline) and since 2000 for diesel engines [1, 2]. Nowadays cars have multiple sensors regarding monitoring mechanical systems, electrical systems and chemical systems (exhaust gas composition).

2 OBD II Connector Pinout and Protocols The OBD II connector is standardized and the connector pinout is shown in the Fig. 1. The location of OBD II connector in the automobile can be in different places, but most common ones are below the A-pillar, or below the steering wheel column or hidden in the middle console or under the passenger seat or in the passenger seat luggage compartment. The connector must be easily accessible, since the technicians has to have access to it during annual MOT vehicle inspection. In Croatia, the technicians mostly use Bosch (BEA 350) interface during the CO2 emission measurement as per “Commission implementing regulation (EU) 2021/392” [3] where CO2 emission data is collected and processed. It is necessary to fulfill technical demands for CO2 measurement in the form of engine water/oil temperature and levels, exhaust system without holes and rust (elimination of secondary air), the original air filter and housing must be present (no intake modification allowed). It is worth mentioning that Malfunction indicator lamp (MIL lamp or CHECK ENGINE lamp) must not be turned on during this process and with OBD monitoring the possible tampering with the car electrical systems can be detected.

Fig. 1. OBD II J1962 port pinout.

Table 1 shows pinout assignment based on the manufacturers. The SAE J1850 VPW protocol is used by Ford and GM. SAE J1850 PWM protocol is used for Ford vehicles and other with the same infrastructure of ECU. The ISO 9141-2 is protocol used for European vehicle made from 2000–2004 year. The ISO 14230-4 is applied in Asian vehicles from 2003 year onward. The ISO 15765-4/SAE J2480 (CAN BUS) is the latest protocol used in new vehicles, since there was agreement in 2008 year regarding the protocol implementation among ECU manufacturers [1, 4]. When accessing the information system (ECU) two types of data can be obtainet. These are either Diagnostic Trouble Codes (DTC), or Parameter ID (PID) values [5]. Typical PID codes are sorted into groups as shown in the Table 2 [5]:

Maintenance of Automobiles and Motorcycles

173

Table 1. OBD II pinout assignment [1] Pin

Description

Pin

Description

1

Manufacturers dicretion

9

Discretionary

2

CAN/SAE J1850 BUS+

10

SAE-J1850 BUS_/PWM only

3

BUS positive of SAE

11

Ford DCL

4

Chassis ground

12

Discretionary

5

Signal ground

13

Discretionary

6

CAN BUS high (ISO 15765-4 and SAE J2284)

14

CAN bus low (ISO 15765-4)

7

K-line of ISO 9141-2 and ISO 14230-4

15

L-line of ISO 9141-2 and ISO 14230-4

8

Discretionary

16

Battery voltage (+)/always on

Table 2. PID modes and explanation [5] PID mode

Explanation

Mode 1

Real time engine data

Mode 2

Fault information that is currently detected in the ECU

Mode 3

Trouble codes stored in the memory

Mode 4

Command for deleting of stored diagnostic trouble codes, and for shutting off warning check engine light

Mode 5

Test results of lambda probe (oxygen sensor)

Mode 6

The rest of test results that are not listed but are continuously monitored

Mode 7

Pending DTC’s (engine is working but some trouble codes come and go intermittently

Mode 8

Special control mode related to manufacturer (like operation of the installed systems)

Mode 9

Read engine VIN and vehicle information

These codes are often generic and experienced technician knows how to address problems stored in the ECU. Some errors appear under same error codes like P0380 (shown in the Fig. 2). When the code P0380 is searched, the solution falls to the: – – – – –

Malfunctioning glow plug or relay Faulty glow plug timer Faulty glow plug module Blown fuse in the circuit Faulty wiring and electrical connections.

174

J. Cumin et al.

This is list of possible errors with the common cause of lack of preheating in the cylinder head, which leads to harder starting of the engine, rough idle, black smoke and violation of the exhaust gas composition and tolerances which can lead to failed M.O.T. inspection. If the Exhaust Gas Recirculation (EGR) valve is dirty, it can furthermore inject dirty exhaust gasses in the intake system and contribute to the black smoke and problematic engine operation [6]. EGR valve is important part of diesel engine and its malfunction code (P0401) is detected by oxygen sensor in the exhaust [7, 8].

Fig. 2. Diagnostic trouble code (DTC) read from the ECU (Toyota Avensis wagon 2.0 D4d 2002 year via ELM327 ODB II module).

3 OBD II Devices As technology advances, a new microchips and devices are produced and the functionality of OBD II technology expands. For the purpose of this article, the average vehicle age is considered. It is expected that new vehicles are under warranty and regularly maintained, and on the opposite end of the spectrum are either oldtimer vehicles or old abandoned vehicle. This indicates that most vehicles on the road (which have average life expectancy) are vehicles that require regular maintenance in order to be in the perfect working conditions. The Croatian national vehicle inspection center (CVH) has issued statistics about this. The average age of personal vehicles (M1 in 2022) was 13,29 years [9]. OBD II is installed in some small portion of motorcycles also, and for that category (L3 in 2022) the average age is 13,54 years [9]. Here shown data is referred to the complete vehicle (not only the malfunctions that can be detected by OBD II technology). The main causes for vehicle defects are:

Maintenance of Automobiles and Motorcycles

175

– Suspension and steering part defects (ball joints, tie-rods, bushings, bearings, broken springs, broken stabilizers, increased clearance in steering rack, defective shock absorbers etc.) – Brake part defects (mostly on mechanical parts but the defect of malfunctioned ABS module can be detected with OBD II) – Tyre defects – Exhaust gas composition abnormalities (too much fuel or oxygen, missing or malfunctioned EGR, DPF, holes in the mufflers). Some errors can be seen via the oxygen sensor (or lambda sensor)/exhaust pressure sensor/exhaust temperature sensor. Other problems such as rusty (holes) exhaust pipe or mufflers are detected at the stage of gas composition measurement with external devices during the M.O.T.

Fig. 3. Percentage of defects in vehicles during the annual M.O.T. inspection sorted by manufacturers (data extracted from CVH annual statistics [9]).

Special problem with vehicle maintenance and M.O.T. compliance to technical regulations was during the COVID 19 pandemic, since the obtainability of the parts was diminished, and it was very difficult to organize repair in mechanical workshops [10]. Here the cheap handheld OBD II devices proved to be useful for DIY repairs [10]. 3.1 ELM 327 The ELM 327 is a chip specially designed to communicate with vehicle ECU and translating on-board error codes, and can be connected by USB cable, RS232, Bluetooth and Wi-Fi. This module work on laptops, tables and phones working on the android devices [11]. Figure 4 shows commercially available ELM 327 Wi-Fi module. This kind of interface modules are among the cheapest solutions – where the price estimate is ranging from 4–17 USD on Ebay (2023/06/19). This module can be connected to retrofitted android-based Hi-Fi devices (android radios) on the vehicles, and in such way the user can monitor live data of the ECU during the ride.

176

J. Cumin et al.

Fig. 4. ELM 327 Wi-Fi interface example.

3.2 K+DCAN/K+CAN, K-line This kind of modules have built in two K-lines and D-CAN, specifically made for BMW vehicles as per Fig. 1 and Table 1. There are also variants of VAG K+CAN+UDS interfaces which can be connected to VW/Audi/Seat/Škoda concern. These kinds of modules are mainly based on the FT232R USB universal asynchronous receiver/transmitter (UART) interface ICs. The datasheet with respective technical characteristics for FT232R can be found online. For the end user it is more important the wider scope of possibilities. The price range estimate is from 50–80 USD. A larger data sets and larger set of options is at disposal to the technicians. For example, the Fig. 5 shows USB cable interface for BMW vehicles.

Fig. 5. D+DCAN USB interface for BMW vehicles.

Maintenance of Automobiles and Motorcycles

177

3.3 Stand-Alone Devices Based on the Standard OBD-II Protocol These devices have basically the same functions as ELM 327 based devices, but they do not require a PC/laptop/tablet/android phone. The internal component structure of these devices varies from manufacturer to manufacturer, and it would be cumbersome to show all the available models. In summary all of these devices can read DTC’s, stored DTC’s, live data etc., just as the ELM 327 devices do. They are widespread in the workshops, easy to use, have a good battery life, resistant to dirt, impact and scratches. Typical representatives are shown in the Fig. 6 (Autel AL319, 529 and 619) [12]. Respective models and their technical characteristics can be found on the internet, and the respective price range is from 30–100 USD. These devices cannot read or modify higher ECU functions, such as customization of vehicle options (“follow me home”, comfort window opening, automatic adjustment of side mirrors when going to reverse gear…), injector coding, ECU map editing, calibration of other mechanical components etc.

Fig. 6. AutoLink AUTEL hand-held devices [12].

3.4 Stand-Alone Devices Based on the Android Systems The example of this kind of diagnostic tool is Launch CRP919X diagnostic tool [13]. It is composed of Android based device with touch screen in a robust housing. This integrated version of a tablet device and OBD II interface is also robust and shock proof and ideal for car shops. Besides standard OBD protocols they can handle new Ethenet based DoIP - Diagnostic over Internet Protocol. This protocol is defined by ISO 14229 standard as Unified Diagnostic Services (UDS), and it combines ISO 14230 (KWP) and diagnostic on CAN (ISO 15765) [14, 15]. In this type of communication the external device has the property of a “client”, and ECU has the property of a “server”, and communication is initiated by the device where this request contains Service ID (SID) and a related subfunction that we want to address and the ECU gives back response [15].

178

J. Cumin et al.

The example of commercially available diagnostic tool is shown in the Fig. 7 where the price range is the highest of the shown devices (estimated 420 USD). But it offers a wider range of data that can be obtained from the car ECU system since it utilizes dedicated port-scanner system [15].

Fig. 7. Launch CRP919X diagnostic tool [13].

3.5 Motorcycle Related Diagnostic Tools When referred to the average age of a motorcycle on the Croatian roads (13,54 years [9]), the equivalent diagnostic tools are shown. The Croatian motorcycle population is mostly dominated by the Japanese motorcycles (Kawasaki, Yamaha, Honda, Suzuki). The off-road motorcycle segment mostly consists of the European brands (BMW, KTM, Triumph, Aprilia). It is difficult to make common denominator among these brands since, for the average age of the motorcycles, there are dominant manufacturers of ECU systems. For Japanese-made motorcycles it is mostly DENSO or Mitsubishi, and for the European motorcycles it is common for the ECU to be produced by Bosch or Siemens. The DENSO ECUs are often protected and cannot be modified, but they have internal diagnostic system which transfers the problem code via LED light when the service connector is connected to the motorcycle ground. Some of ECU modules are made for racing, and there are solutions for connectivity between a PC and ECU (such as Woolich racing ECU adapter). Through this, the users (technicians) can read trouble codes but it is not based on the OBD II technology. In addition to maintenance part, the Woolich racing interface can read and edit ECU’s fuel maps and ignition maps [16]. There are universal diagnostic devices such as OBDPROG MOTO 100 EU [17] which can be used for the maintenance of motorcycles form BMW, Ducati, KTM, Honda, Yamaha and Triumph. There is also a software solution “Tune ECU” for the same brands, which utilize a standard OBD II diagnostic device (for example form Fig. 4) and it can be found at [18]. For the Bosch made ECUs installed in BMW motorcycles there is a specific hardware solution called HEX GS-911 which can read pending and stored DTCs, read sensor data, activate output, make basic coding etc. [19].

Maintenance of Automobiles and Motorcycles

179

4 Application of OBD II Diagnostic Device in the Field of Maintenance Every vehicle has a recommended time sheet for the preventive maintenance, where the technician can find operation description, service time, next service interval etc. This time sheet can look like in the example shown it Table 3. Table 3. Example of preventive maintenance Action

0 km

Every 3000 km

Check tyre pressure

•

•

Check oil level

•

•

Check brake oil level

•

•

Inspect fuel filter

Every 10000 km

After 50 spend fuel tanks

Every oil change

•

•

These service recommendations can be found in the web pages or in the specialized software such as AUTO-DATA. Some of the items can be easily checked visually, some of them need to be diagnosed with special equipment. The usage of diagnostic tools is very helpful to pinpoint the failed component or assembly without need of measuring data in sight. When the trouble code (DTC) is obtained, for each vehicle there is a recommended set of actions. Some actions require small effort, while others require extensive search for the problem source. There is a misconception that OBD devices can measure sensor data directly, or that some components can be directly actuated. Instead the commands are given through shown protocols and devices, and if there are communication problems they can be traced. For example: if the Mass Air Flow sensor signal is missing in the ECU the CHECK ENGINE will be lit, and ECU will compensate for the value of the sensor (the engine will likely be in the limp mode). So, technician can see the missing sensor signal problem and can measure voltage directly at the sensor (find out if it is dirty or defective), or there is problem of broken wire(s) in the wire loom.

180

J. Cumin et al.

As another example, if there is malfunction with the fuel pump – the engine will work erratically, or in the worst-case scenario it will not start at all. With the diagnostic tool the technician will most likely obtain P0230 (Fuel Pump Primary Circuit Malfunction) code. After this he can follow table of recommended actions from the service manual, but he can also try to activate fuel pump relay through diagnostic tool. In this case, it is easy to inspect whether the fuel pump relay is operating properly, or if there is problem with fuel pump fuse. If everything is working in order, then he can address the fuel pump itself (electrical motor). Some malfunctions can be detected when the engine is working, for example the position of camshafts in variable timing systems can be read and technican can see whether they are in the required position or there is some sort of a problem which can be observed during variable engine RPM. Similar case is the ECU correction of short-term fuel trim (STFT) and long-term fuel trim (LTFT) in petrol engines, or injector mass adaptations in diesel engine. In both cases the live data measuring obtained by OBD diagnostic tool proves to be irreplaceable asset in the mechanical workshop. The android devices with integrated interface and touch screen are the best choice for workshops, but how can one make a criterion for purchase of the optimal tool? Most of the newer vehicles produced by the EU manufacturers have implemented OBD II technology with various “comfort” coding functions, and as an extreme example it can be seen that the BMW even planned to charge customers monthly fee for heated seats and driving assistance technologies [20]. So this level of coding has to have a good backup in hardware and it can only be assumed that this kind of coding equipment will not ne available to average mechanic (or at least it can be assumed that maybe his level of informatic knowledge will not be up to the task of performing such maintenance). For the end customer (if the owner does his own maintenance) it is a matter of finding optimal budget for the OBD II devices, and that is heavily based on the price and second on the coding options. “Low level” technical devices (such as Fig. 4, Fig. 5) can satisfy most of the maintenance requirements for moderately priced vehicles, such as shown in the CVH statistic data from Fig. 3 [9]. For the professional use, there are hardware solution based on the OBD II protocol (ENET, DoIP) which can access even hidden data such as engine fuel and ignition maps, vehicle mileage (the existing problem of odometer rollback). The price of such solutions is extreme for home usage, and the technical documentation is vast, so only a highly trained technicians can operate these devices like Microtronic Autohex (price starting at 2300 USD) [21]. Table 4 shows a group data of the technical features for the devices mentioned in the text. These are only main features, and often such devices come with an elaborate User manuals and/or technical backup on the internet, so it is impossible task to show all the functions in the comparison table.

Maintenance of Automobiles and Motorcycles

181

Table 4. Comparison of technical data for the OBD modules ELM 327

K+DCAN (FT232R)

AUTEL

CRP919X

AUTOHEX II

Protcols

SAE J1850 PWM SAE J1850 VPW ISO 9141-2 ISO 14230-4 (KWP) ISO 15765-4 (CAN) SAE J1939 SAE J1939

Same as ELM 327 but with option for bridging pins 7&8 (for the K-line), or leaving pins 7&8 separated (DCAN)

SAE J1850 PWM SAE J1850 VPW ISO 9141-2 ISO 14230-4 (KWP) ISO 15765-4 (CAN) SAE J1939 SAE J1939

SAE J1850 SAE J1850 ISO 9141-2 ISO 14230-4 (KWP) ISO 15765-4 (CAN) SAE J1939 SAE J1939

SAE J1850 SAE J1850 ISO 9141-2 ISO 14230-4 (KWP) ISO 15765-4 (CAN) SAE J1939 SAE J1939

Coding options

- Read DTC - Clear DTC - Monitor real-time sensor data - read coolant temperature - read AIT (intake temperature) - read MAF airflow data - Read Throttle Position Sensor data (TPS) - Read oxygen (lambda sensors) data

- Read DTC - Clear DTC - Monitor real-time sensor data - read coolant temperature - read AIT - read MAF - Read TPS - Read O2 data - Read and write different ECUs: (Car Access System CAS, Dynamic Stability Control – DSC, Parking Distance Control – PDC, Digital Motor Electronics – DME, Integrated Heating and Air conditioning – IHKA and so on

- Read DTC - Clear DTC - Monitor real-time sensor data - read coolant temperature - read AIT - read MAF - Read TPS - Read O2 sensordata

- Read DTC - Clear DTC - Monitor real-time sensor data - read coolant temperature - read AIT - read MAF - Read TPS - Read O2 sensordata - Read and write different ECUs CAS DSC PDC DME IHKA

- Read DTC - Clear DTC - Monitor real-time sensor data - read coolant temperature - read AIT - read MAF - Read TPS - Read O2 sensordata - Read and write different ECUs CAS DSC PDC DME IHKA Read and write engine maps Read and write Odometer information

Stand alone

No

No

Yes

Yes

Yes

Price

4 ÷ 17 USD

50 ÷ 80 USD

30 ÷ 100 USD

420

2300 SD

5 Conclusion This paper gives an overview of the basis of On-Board Diagnostic (OBD) technologies. The fist section of the paper gives vehicle standardized OBD II port pinout. It is shown that assigned ODB II port pins work on different communication protocols, and different communication speeds. Standardized Parameter IDs are shown with respective modes. The standard functions are also described. In the rest of the paper, the technical data of some commercially available ODB II devices is shown with some of the respective features in effort to give brief insight into technological possibilities.

182

J. Cumin et al.

As the technology is thriving, it is imperative to keep track of the vehicle data (either infotainment data, or engine data). For this reason, a new communication chips and software solutions are developed each day. In order to maximize vehicle life it is important to take care of the regular maintenance. This maintenance is mechanical (preventive and corrective), and electrical maintenance. Modern vehicle hardware components are filled with electronic devices and some hardware components require new firmware flashing before replacing the damaged (old) components or retrofitting from one car series to the other car series (it often requires flashing of correct data parameters). Some devices are “all in one” (hand-held), while other require some sort of external computer to work (windows or linux based personal computers/laptops, or android devices). The owner of the vehicle (DIY maintenance), or experienced technicians can search through detailed technical characteristics of the devices and of the existing software solutions and make buying choice based on the best buy ratio. If average vehicle data in Croatian roads [9] is taken into maintenance perspective, it can be concluded that the shown cheapest solutions (Android smartphone+ELM 327; or Autel 619) can satisfy basic needs. These devices give general PID codes and point in the general direction of malfunction, where the system of elimination, or the existing repair manual troubleshooting tables help in repair. Some middle and higher class vehicles have many electronic circuits and systems and require special communication protocols among vehicle ECU and a OBD II software. Here the OBD II communication devices are somewhat expensive and require substantial computer and programming skills and they are intended for maintenance tehnicians. The modern solutions like AUTOHEX II give the most possibilities in reading and writing ECU data. In this way, not only the standard errors can be read and cleared, this device can access ECU fuel and ignition MAP and change its values. This is often used by engine tuners to modify Air/Fuel (A/F) ration for best performance, but it also means that this vehicle is most likely not compliant with the homologation data. These kind of devices can easily read odometer values and write another one (which is useful if the instrument cluster or ECU breaks and need to be replaced with another one), but instead this function is often badly used for odometer rollback scam.

References 1. Ramai, C., Ramnarine, V., Ramharack, S., Bahadoorsingh, S., Sharma, C.: Framework for building low-cost OBD-II data-logging systems for battery electric vehicles. Vehicles (MDPI) 4(4), 1209–1222 (2022). https://doi.org/10.3390/vehicles4040064 2. Aris, I., Zakaria, M.F., Abdullah, M.F., Sidek, R.M.: development of OBD-II driver information system. In: International Engineering Convention, pp. 108–114. FEIIC, Jeddah (2007) 3. Commission implementing regulation (EU) 2021/392, Official Journal of the European Union. https://eur-lex.europa.eu/legal-content/HR/TXT/?uri=CELEX:32021R0392. Accessed 02 June 2023 4. El-Den, B.M., Mohamed, M.A., AbdelFattah, A.I.: Safe vehicle driving using android based smartphones. In: Abraham, A., Jiang, X., Snášel, V., Pan, J.S. (eds.) Intelligent Data Analysis and Applications. Advances in Intelligent Systems and Computing, vol. 370, pp. 291–303. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-21206-7_25

Maintenance of Automobiles and Motorcycles

183

5. Rimpas, D., Papadakis, A., Samarakou, M.: OBD-II sensor diagnostics for monitoring vehicle operation and consumption. Energy Rep. 6, 55–63 (2020) 6. Hochgreb, S.: Chapter 6 - Combustion-related emissions in SI engines. In: Handbook of Air Pollution from Internal Combustion Engines, pp. 118–170. Academic Press, New York (1998) 7. Kumar, M., Cramsky, J., Lowe, W., Danielsson, P.O.: A prediction model for exhaust gas regeneration (EGR) clogging using offline and online machine learning. In: Berns, K., Dressler, K., Kalmar, R., Stephan, N., Teutsch, R., Thul, M. (eds.) ICVTS 2022, pp 185–198. Springer, Wiesbaden (2022). https://doi.org/10.1007/978-3-658-40783-4_13 8. Delphi Technologies by Borg Warner: The basics of EGRs. https://www.delphiautoparts. com/en-gb/resource-center/article/the-basics-of-egrs---what-they-do-how-they-work-howto-troubleshoot. Accessed 16 June 2023 9. CVH statistic – open access. https://www.cvh.hr/gradani/tehnicki-pregled/statistika/. Accessed 19 June 2023 10. Glavas, H., Karakasic, M., Kljajin, M., Desnica, E.: Essential preventive automobile maintenance during a pandemic. Tech. Gazette 28, 2190–2199 (2021) 11. ELM 327. https://www.elm327.com/plus/list.php?tid=5. Accessed 19 June 2023 12. AutoLink 319. https://www.autel.com/mk1/3239.jhtml. Accessed 19 June 2023 13. LAUNCH Scan Tool CRP919X 2023. https://www.launchx431pro.com/products/launchscan-tool-crp919x. Accessed 19 June 2023 14. Weiss, N., Renner, S., Mottok, J., Matoušek, V.: Automated threat evaluation of automotive diagnostic protocols. In: 19th hybrid ESCAR conference, pp 1–20. ESCAR (2021) 15. Sommer, F., Durrwang, J., Wolf, M., Juraschek, H., Ranert, R., Kriesten, R.: Automotive network protocol detection for supporting penetration testing. In: SECURWARE 2019: The Thirteenth International Conference on Emerging Security Information, Systems and Technologies, pp. 144–119. IARIA (2019) 16. Woolich racing. https://www.woolichracing.eu/products. Accessed 20 June 2023 17. OBDPROG MOTO 100 EU. https://www.obdprog.com/product/Detection%20Tool/2771. html. Accessed 19 June 2023 18. TUNE ECU. https://tuneecu.net. Accessed 20 June 2023 19. HEX 911 GS. https://www.hexgs911.com/product/. Accessed 20 June 2023 20. Euronews. https://www.euronews.com/next/2022/07/14/bmw-charging-a-monthly-fee-forheated-seats-is-a-glimpse-into-the-future-of-cars. Accessed 20 June 2023 21. Autohex II diagnostic scan tool. https://www.microtronik.com/products/autohex-ii-scan-tool. Accessed 20 June 2023

Construction of Variable Sheet Metal Hand Bending Tool Josip Cumin1(B)

, Hrvoje Vorel1 , Miroslav Duspara1

, and Hrvoje Glavaš2

1 Mechanical Engineering Faculty (MEF - SFSB), University of Slavonski Brod (UNISB), Trg.

I.B. Mazuranic 2, 35000 Slavonski Brod, Croatia [email protected] 2 Faculty of Electrical Engineering, Computer Science and Information Technology (FERIT), Kneza Trpimira 2B, 31000 Osijek, Croatia

Abstract. Automotive chassis is made from deep drawn and stamped sheet metal pieces. Corrosive resistance of the chassis is maintained by the means of galvanizing technology in synchronization with the usage of modern coatings. During time, the coating can be chipped off - especially on the chassis underbody (e.g. from gravel or rocks). In such places corrosion quickly develops and the sheet metal deteriorates. This chassis needs to be repaired in order to prolong vehicle lifetime. The body repair is performed by cutting-out the corroded segments and replacing it by welding. For the modern cars, the sheet metal parts are easily obtainable in the specialized stores, but there is problem of finding sheet metal parts for “oldtimer” vehicles. In such cases, the sheet metal parts are required to be custom tailored with the basic tools and excellent problem-solving and manufacturing skills. This paper brings the idea of adjustable (modular) hand-bending tool for the bending of sheet metal segments, so that they could be custom tailored to the car chassis damaged place. Keywords: bending · tool · construction · sheet metal

1 Introduction This paper deals with the construction and dimensioning of working elements of a hand-operated sheet metal bending tool. One of the applications of this bending tool is in the sheet metal automotive (body) workshops. Croatia is located in the southeast part of the Europe, thus having geographical diversity which is related with cold and snowy winters, and mild summers on the seaside. During winter, the inner part of the Croatia is often under influence of cold weather, and it is usual for road maintenance crews to use salt for the prevention of black ice on the roads. Similarly, on the seacoast there are powerful wind blows which can carry salty seawater. For the abovementioned reasons, the car chassis is exposed to the corrosive environment and prone to the rusting process. During the annual M.O.T vehicle inspection, one of the parameters for the road safety is the lack of corrosion on body parts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 184–198, 2024. https://doi.org/10.1007/978-3-031-51494-4_17

Construction of Variable Sheet Metal Hand Bending Tool

185

Most cars nowadays have chassis produced from the stamped steel body panels which were welded together, and with the application of some sort of anti-corrosion technical process [1]. Usually the chassis is galvanized with the zinc layer or with some other element which acts like corrosion inhibitor, such as phosphating process [1]. In some countries, the winter maintenance crews use fine gravel for the roads, and this gravel can damage the chassis coatings, thus leaving bare metal exposed to the corrosive environment. In due time, these parts can be totally corroded and problematic for the road safety. It is usual, for the newer automobiles to have the replacement car body panels easily obtainable, and the maintenance of the car body can be performed. The problem of repair exists for the youngtimer/oldtimer vehicles, for which body panels are unobtainable and during repair/maintenance only the sheet metal segments of body panels need to be manufactured by hands of the experienced craftsmen. Here shown hand-operated bending tool is usually used for such purposes. 1.1 Basics of Sheet Metal Bending In the literature, the sheet metal bending theory can be found (in great details). For the construction of sheet metal bending tool, some basic expressions are given, with respect to the material properties. Figure 1 shows a plane strain bent segment of a sheet metal, with respective parameters. Rn – represent the neutral axis radius/mm, Ri – inner radius/mm, Ro – outer radius/mm, s- sheet thickness/mm, ϑ - bending angle/radian.

Fig. 1. Sheet metal segment bent with pure bending moment M.

186

J. Cumin et al.

On the right side of the Fig. 1, the bend stresses can be observed in a simplification of material behavior which is often used for quick assessment of bending moment M [1, 2]. Material can be assumed to behave in linear elastic–plastic mode where outer fiber is under tension stress and inner fiber under compression stress. Or the calculation of the necessary pure bending moment M can be found by the utilization of pure plastic behavior of the sheet metal and it is achieved for Rn /s < 5. Approximation of bending moment by linear elastic – plastic theory is given by Eq. (1), where Rp0.2 is yield strength, b – sheet width/mm [1]: M = Rp0.2 ·

b · s2 /N · mm 4

(1)

Approximated bending moment in pure plastic bending theory can be determined [1, 3]: M = kf ·

b · s2 /N · mm 4

(2)

where k f – true stress of the material/MPa. There are theories which describe bending physics in more detail [4–6] but for the calculation of bending moment for the constructed bending tool - these simplified theories (1, 2) are used under assumption that the neutral line stays in the same position during the bending (Rn = Ri + 0.5s from Fig. 1), which seldom appears in practice. The simplified calculation of the neutral axis movement during bending was given by Hill, and it can be used for the calculation of bend allowance [1]:  Rn = R0 · Ri /mm (3) Material strains of the inner and outer fibers are important to calculate. If the outer fiber stretches too much (tensile stress), there is the risk of surface cracking. Calculation of the strains is usual to be in the form of an engineering strain ε (4) or truestrain ϕ o (5), where ϑ is the bending angle in radians:   s Rn + 2s · ϑ − Rn · ϑ l o − ln = 2 = (4) ε= ln Rn · ϑ Rn   Rn + 2s (5) ϕo = ln Rn

Construction of Variable Sheet Metal Hand Bending Tool

187

2 Materials For the construction of bending tool, the material mechanical properties need to be considered. In the automotive industry a large variety of materials are used. Materials with high formability, such as HC260Y and similar are used in the large deformation zones (panels that intentionally buckle during the collision). DC04 is material with good formability generally used for deep drawn products. Inoxidable materials are also used, for e.g. AISI 321 (X6CrNiTi18-10). Some of these materials with sheet thicknesses up to 3 mm are selected for the calculation of bending tool, and their mechanical properties are shown in the Table 1. Table 1. Selected sheet metal materials related to automotive industry Material:

EN

Yield strength Rp0.2 /MPa

Tensile strength Rm /MPa

Maximal elongation A/%

True stress vs. true strain

S355J2 [7]

1.0577

355

490–630

22

Equation (6)

HC260Y [8]

1.0928

260–320

380–440

31

Equation (7)

DC04 [9]

1.0338

210–220

270–350

37–38

Equation (8)

X6CrNiTi18-10

1.4541

190–225

810–850

35

Equation (9)

The true stress-true strain curves were obtained by the literature and are valid for temperature range of 20–180 °C, and true strain range ϕ = 0.03 ÷ 2: a) S355J2 [10]: kf = 843.34 · ϕ 0.12479 · e−0.00736/ϕ /MPa

(6)

b) HC260Y [11] (ϕ = 0.03 ÷ 0.35): kf = 683.88(0.003 + ϕ)0.2267 /MPa

(7)

kf = 368.34 · ϕ 0.14496 · e0.00013/ϕ /MPa

(8)

kf = 1345.36 · ϕ 0.37722 · e−0.0231/ϕ /MPa

(9)

c) DC04 [10]:

d) X6CrNiTi18–10 [10]:

From Eq. (5), the minimal bending radius Ri /mm, in relation to material thickness can be calculated: s(1 − ε) /mm (10) Ri = 2ε Table 2 shows calculated values related to specific bending moment (per 100 mm sheet width), and minimal bending radii (Ri ) for S355J2. Table 3 shows the same parameters but for HC260Y steel, Table 4 for DC04 steel and Table 5 for X6CrNiTi18-10 steel.

188

J. Cumin et al. Table 2. Bending parameter data for S355J2 steel based on expressions (2–10).

Sheet thickness s/mm

Minimal radius Ri /mm

Geom. Outer radius Ro /mm

Geom. Neutral axis Rn /mm

Neutral axis (by Hill) Rn /mm

True strain ϕ

True stress fiber k f /MPa

Bend. Moment (100 mm) width, Nm

0.75

1.33

2.08

1.7

1.66

0.1988

664.9

9.3

1

1.77

2.77

2.27

2.22

0.1988

664.9

16.6

1.5

2.66

4.16

3.41

3.33

0.1988

664.9

37.4

2

3.55

5.55

4.55

4.43

0.1988

664.9

66.5

3

5.32

8.32

6.82

6.65

0.1988

664.9

149.6

Table 3. Bending parameter data for HC260Y steel based on expressions (2–10). Sheet thickness s/mm

Minimal radius Ri /mm

Geom. Outer radius Ro /mm

Geom. Neutral axis Rn /mm

Neutral axis (by Hill) Rn /mm

True strain ϕ

True stress fiber k f /MPa

Bending moment M/N·m

0.75

0.83

1.58

1.21

1.15

0.27

697.4

9.8

1

1.1

2.11

1.61

1.53

0.27

697.4

17.4

1.5

1.67

3.17

2.42

2.3

0.27

697.4

39.2

2

2.23

4.23

3.23

3.07

0.27

697.4

69.7

3

3.34

6.34

4.84

4.6

0.27

697.4

156.9

Table 4. Bending parameter data for DC04 steel based on expressions (2–10). Sheet thickness s/mm

Minimal radius Ri /mm

Geom. Outer radius Ro /mm

Geom. Neutral axis Rn /mm

Neutral axis (by Hill) Rn /mm

True strain ϕ

True stress fiber k f /MPa

Bending moment M/N·m

0.75

0.63

1.38

1

0.93

0.3185

714.8

10.1

1

0.83

1.83

1.33

1.24

0.3185

714.8

17.9

1.5

1.25

2.75

2

1.85

0.3185

714.8

40.2

2

1.67

3.67

2.67

2.47

0.3185

714.8

71.5

3

2.5

5.5

4

3.71

0.3185

714.8

160.8

Construction of Variable Sheet Metal Hand Bending Tool

189

Table 5. Bending parameter data for X6CrNiTi18-10 steel based on expressions (2–10). Sheet thickness s/mm

Minimal radius Ri /mm

Geom. Outer radius Ro /mm

Geom. Neutral axis Rn /mm

Neutral axis (by Hill) Rn /mm

True strain ϕ

True stress fiber k f /MPa

Bending moment M/N·m

0.75

0.7

1.45

1.07

1

0.3001

708.6

10

1

0.93

1.93

1.43

1.34

0.3001

708.6

17.7

1.5

1.39

2.89

2.14

2.01

0.3001

708.6

39.9

2

1.86

3.86

2.86

2.68

0.3001

708.6

70.9

3

2.79

5.79

4.29

4.01

0.3001

708.6

159.4

3 Tool Construction The sheet metal bender construction is shown in the Fig. 2. Its fixed base is made out of UNP 120 mm ISO 657/11-1980(E) segment, with welded sides made out of L-profile L100 × 50 × 10; (ISO 657), which is shown in the Fig. 3.

Fig. 2. Sheet metal hand bending tool assembly.

Figure 4 shows the upper clamping part of the bending tool with adjustable wiper and the rest of the assembly which is used for maintaining tool rigidity. It is made out of the L50 × 50 × 6 (ISO 657-1) profile, (ISO1035-3) plate 510 × 75 × 6 mm, pipe

190

J. Cumin et al.

Fig. 3. Base (fixed part) of the sheet metal bending tool.

segments of φ50 × 5 mm. The wiper of the clamping bar is held in place with four M8 × 16 (ISO4762) screws, and it has a radius of 1 mm and angle of 45° which allows sheet metal bending angle of 145° in total (shown in the Fig. 5).

Fig. 4. Movable part of the bending tool with respective parts.

Figure 6 shows that the upper clamping part of the bending tool is rested on the springs, which in turn lifts the whole clamping assembly. This is useful for easy sheet metal manipulation and positioning. The tightening of screws by hand is user friendly feature, and it is provided by M16 × 1,5 mm screw specially designed for this purpose (Fig. 7).

Construction of Variable Sheet Metal Hand Bending Tool

191

Wiper angle of 45° and radius of 1 mm

Fig. 5. Adjustable wiper position.

Fig. 6. The upper clamping assembly of the bending tool is floating on the springs which benefits sheet metal manipulation, positioning and clamping.

In order to check stresses in working parts of the bending tool (in linear-elastic mode), only selected parts of the assembly were subjected to the Finite Element Model (FEM) analysis, based on the required estimated amount of bending moment.

4 FEM Numerical Simulation From Fig. 2, it can be seen that the lever length is chosen as 300 mm in order to keep bending tool compact and ergonomically adapted to physiognomy of average worker height of 180 cm and weight of 75 kg. Thus, the maximal possible lifting force that the worker can apply is selected as 500 N (this force can easily be achieved by using another

192

J. Cumin et al.

Fig. 7. Tightening screw M16 × 1.5 specially designed for easy manipulation.

lever). So the tool components and maximal sheet metal width b are selected/designed with respect to the applied working moment of M = 500·0.3 = 150 N·m. The Table 6 gives maximal sheet width with respect to the material properties from Tables 2, 3, 4 and 5. This presents technical data or working parameters of the bending tool. The width of flying bending element is 516 mm, which is shown in the Fig. 8, which gives insight into applicable bending process parameters window. Table 6. Maximal sheet metal width that can be bent with 150 N·m of moment Sheet thickness/mm

Maximal sheet metal width b/mm S355J2

HC260Y

DC04

X6CrNiTi18-10

0.75

1505

1529

1492

1505

1

846

860

839

846

1.5

376

382

373

376

2

211

215

209

211

3

94

95

93

94

From the data shown in Table 6, the width b of the thickest sheet metal is interesting for tool construction. Although this sheet metal thickness is small (b = 93 mm), it is important because if the sheet metal plate is not placed symmetrically in the tool, during bending operation – the tool is loaded unfavorably, and it can lead to uneven elastic deformation of the tool. Figure 9 shows linear elastic analysis for the lower tool part, for the case of nonsymmetric bending of a 94 mm wide strip sheet metal, and with the 3 mm thickness. Figure 10 shows stresses from the bottom side of the flying element, and Figs. 11 and 12

Construction of Variable Sheet Metal Hand Bending Tool

193

Fig. 8. The measured width of flying bending element.

show displacement of the lower tool under loading condition (non-symmetric loading case).

Fig. 9. Lower tool stresses for bending a strip of 3 mm thick and 94 mm wide.

194

J. Cumin et al.

Fig. 10. Stresses from the bottom side of the flying element.

Fig. 11. Displacements of the lower part of bending tool under non-symmetrical loading conditions.

Construction of Variable Sheet Metal Hand Bending Tool

195

For the mild steel material properties (S235), the yielding stress is Rp0.2 = 235 MPa, and it can be concluded that with regard to obtained stresses from Fig. 9, the bending tool construction is acceptable and it shouldn’t be damaged during bending (operated by hands). The calculated displacements from Fig. 11 are also low, and the do not have effect on the bent sheet metal geometry, nor on the lower bending tool assembly since these displacements and respective deformations are well in the linear-elastic domain. Figure 12 shows calculated stresses of the lower bending tool for the symmetric loaded sheet metal, and Fig. 13 shows displacements which are also low and well in elastic domain.

Fig. 12. Equivalent Von Mises stresses for the symmetrical loading of sheet metal in the bending tool.

Figure 14 shows stresses in the isolated segment of a hinge element, which are also in linear-elastic domain, and it is indicated that this lower flying element of the bending tool is also properly dimensioned. The hinges are designed to have grease nipples in order to minimize wear of the contact surfaces and for prolonged tool life (shown in Fig. 15).

196

J. Cumin et al.

Fig. 13. Displacements in the bending tool for symmetrical sheet metal loading.

Fig. 14. Von Mises stresses in the hinge element.

Construction of Variable Sheet Metal Hand Bending Tool

197

Fig. 15. Hinges and drilled stud connector for grease nipple and grease passages.

5 Conclusion This paper deals with the design and construction of hand operated sheet metal bending press. This kind of tool is often used in the mechanical workshops for the hand tailoring and bending of sheet metal automotive patches/parts. There is variety of alike bending tools. The bending tool showed here was designed with respect to small mass, ease of attachment to the workbench, good working conditions (human ergonomics), and with respect to the tool longevity (stud bolts with grease gallery holes and grease nipples). For the selected materials often used in the automotive industry, the mechanical properties and material flow curves are given. Based on the mechanical properties, the bending moment (torque) is calculated by usage of linear-elastic and pure plastic bending theory. Based on the sheet metal thickness, the minimal bending radii are given, from which the bending allowance can be calculated, which is important for sheet metal bend deduction. The added value of this bending tool is the sheet metal pressure plate/wiper position adjustment in order to accommodate bending parameters for different sheet metal thicknesses, and different bending radiuses. The upper assembly of the bending tool rests on the springs in order to provide easier sheet metal positioning, and the clamping is made with the usage of fine metric thread screws.

References 1. Balasubramanian, J., Kumar, V., Kirubakaran, M., Lalwani, R.: A study on automotive sheetmetal surface pretreatment: liquid activation and low temperature phosphating. In: International Conference on Automotive Materials and Manufacturing AMM 2023. SAE International, USA (2023) 2. Grizelj, B.: Oblikovanje lima deformiranjem (Sheet metal forming). Strojarski fakultet Sveuˇcilišta u Slavonskom Brodu, Slavonski Brod (2009) 3. Siegert, K.: Blechumformung, Verfahren, Werkzeuge und Maschinen. Springer, Heidelberg (2015)

198

J. Cumin et al.

4. Mehrbai, H., Yang, R.: Effects of tension–compression asymmetry on bending of steels. Appl. Sci. 10(9), 3339 (2020) 5. Nie, D., Lu, Z., Zhang, K.: Hot V-bending behavior of pre-deformed pure titanium sheet assisted by electrical heating. Int. J. Adv. Manuf. Technol. 94, 163–174 (2018) 6. Alexandrov, S., Wang, Y.C., Lang, L.: A theory of elastic/plastic plane strain pure bending of FGM sheets at large strain. Materials 456(12), 1–17 (2019) 7. Ovako STEEL NAVIGATOR. https://steelnavigator.ovako.com/steel-grades/s355j2/. Accessed 26 June 2023 8. Thyssenkrupp materials processing Europe. https://www.thyssenkrupp-materials-proces sing-europe.com/en/c-steel/cold-rolled-sheet/high-and-higher-strength-steel/high-strengthif-steel-or-hc180y-hc260y. Accessed 26 June 2023 9. European Steel and Alloy Grades/Numbers Searchable Database. http://www.steelnumber. com/en/steel_composition_eu.php?name_id=199. Accessed 26 June 2023 10. Spittel, M., Spittel, T.: Numerical Data and Functional Relationships in Science and Technology, Group VIII: Advanced Materials and Technologies, vol. 2 Subvolume © - Metal Forming Data Materials. Landolt-Börnstein. Springer, Heidelberg (2009) 11. Cumin, J., Stoi´c, A., Duspara, M.: Bending accuracy of the HC260Y steel in different V-tool configurations. Tech. Gazette 23(1), 229–236 (2016)

Cost-Effectiveness of an Automatic Lubrication System for Bearings Serkan Yildiz1

, Murat Apakhan1

, and Muharrem Hilmi Aksoy2(B)

1 IMAS R&D Center, Mechanical Engineer, 4200 Konya, Turkey 2 Department of Mechanical Engineering, Faculty of Engineering and Natural Sciences, Konya

Technical University, 4200 Konya, Turkey [email protected]

Abstract. Improper lubrication is a leading cause of bearing failures, accounting for half of all instances. Lubricating bearings with grease or oil forms a protective film that prevents direct metal-to-metal contact, reducing friction and overheating, and prolonging the bearing’s lifespan. Lubrication also acts as a barrier against foreign particles and wear. In this study, an analysis of automatic lubrication in a milling plant was carried out and compared to manuel lubrication. A milling factory with 22 roller mills with a total production capacity of 450 tons/day is considered. It is calculated that the failure of eight bearings in each mill roller due to lubrication issues results in a substantial cost of $39,000 over a two-year period. The Automatic lubrication system named “SmartLub” regulates lubricant quantity, timing, and application points, ensuring optimal lubrication. The choice of the oil pump was determined by considering both the viscosity of the oil and the head loss within the piping system utilized. By adhering to calculated frequencies, the system extends bearing service life to four years. Furthermore, the manual system reduces labor costs by $400 per roller mill every two years, while unplanned downtimes caused by lubrication issues are minimized. The automatic system eliminates the need for bi-monthly one-day shutdowns in the conventional lubrication systems, saving a total of six days per year. With an estimated lifespan of 20 years, the system achieved a payback period of 1.23 years, demonstrating its cost-effectiveness and long-term benefits. Overall, automation in bearing lubrication enhances machine efficiency, reduces spare parts and maintenance costs, and ensures optimal lubrication in the grain milling sector. Its implementation leads to extended bearing lifespan, reduced downtime, and improved profitability. Keywords: Lubrication · bearing · economic analysis · milling · maintenance · automation

Nomenclature   E N /m2 : hmin [m]: U: G: W:

  E = E/ 1 − (1/m)2 , Minimum lubricant film thickness in the area of rolling contact Speed parameter Material parameter Load parameter for line contact

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 199–209, 2024. https://doi.org/10.1007/978-3-031-51494-4_18

200

S. Yildiz et al.

k:  a m2 /N : n0 [Pa × s]: v[m/s]: Rr [m]: Rr = r1 × r2 /(r1 + r2 ) Rr = r1 × r2 /(r1 − r2 ) r1 = r2 = Q[N ]: L[m]:

k = a/b, ratio of the sermiaxes of the contact areas pressure-viscosity coefficient dynamic viscosity v = (v1 + v2 )/2, mean rolling velocity reduced curvature Radius at linner ring contact at linner ring contact Rolling element Radius [m] Radius of linner and outer ring raceways [m] Rolling element load Gap length or effective roller length

1 Introduction In the domain of mechanical engineering, industrial machinery and equipment depend on a multitude of moving components to operate efficiently. To guarantee peak performance and durability, routine maintenance and the application of suitable lubricants are indispensable. This paper is dedicated to underscore the importance of bearing lubrication and its consequential influence on machine performance, with particular emphasis on bearings. Figure 1 illustrates the manual bearing lubrication process, portraying a method conducted by operators in the absence of standardized protocols or procedures. 1.1 Bearings Bearings play a crucial role in facilitating the rotational movement and load-bearing capabilities of machine components. To ensure optimal functionality, it is essential to shield bearings from detrimental factors such as friction, heat generation, and wear. 1.2 Functions of Lubrication in Rolling Bearings Lubrication is a critical aspect of both rolling and sliding bearings, aiming to minimize direct contact between surfaces and subsequently reduce friction and wear. In rolling bearings, the primary goal is to create a lubricating film, referred to as “physical lubrication,” by applying oil to the bearing components involved in rolling contact. Although rolling is the primary mode of contact, some degree of sliding occurs due to elastic deformation and functional surface profiles. In cases of pure sliding contact, such as between rolling elements and cages or roller faces and lip surfaces, contact pressure is typically lower compared to rolling conditions. Remarkably, rolling bearings exhibit minimal energy losses due to friction and wear even in challenging lubrication scenarios. This adaptability allows for effective lubrication using greases of varying consistencies and oils with different viscosities, enabling smooth operation across diverse speed and load ranges. In situations where the lubricant film doesn’t entirely separate contact surfaces, low-wear operation is still attainable. Localized temperature increases trigger chemical reactions between lubricant additives and rolling element or ring surfaces, resulting in

Cost-Effectiveness of an Automatic Lubrication System

201

Fig. 1. Manual lubrication of a bearing.

“chemical lubrication.“ Moreover, the efficacy of lubrication can be further improved by incorporating dry lubricants into oil or grease or by employing grease thickeners. In specific cases, rolling bearings can rely exclusively on dry or solid lubricants. These lubricants also serve supplementary functions, including corrosion protection, heat dissipation (in the case of oil lubrication), removal of wear particles and contaminants (via oil circulation lubrication with filtration), and enhancement of bearing seal efficiency (e.g., grease collars, oil-air lubrication) [1]. Accurate lubrication is paramount for optimal bearing performance and serves several critical roles in mechanical engineering: a) Friction Mitigation: Lubricating oil reduces friction between internal bearing surfaces, resulting in decreased energy losses and reduced power consumption. b) Wear and Protection: The presence of lubricating oil acts as a protective layer, guarding against wear, deformation, and damage, especially under sustained loads and high-pressure conditions. c) Thermal Management: Friction during bearing operation generates heat, and lubricating oil plays a crucial role in dissipating this heat, preventing excessive temperature buildup that could compromise bearing integrity. d) Contamination Prevention and Protection: Lubrication effectively removes foreign materials, dirt, and particles that might infiltrate the bearing. It also serves as a protective barrier, shielding the bearing from environmental damage and acting as a safeguard against external contaminants. By effectively fulfilling these functions, precise lubrication practices significantly contribute to the longevity, efficiency, and reliability of bearings in mechanical systems. 1.3 The Influence of Lubrication on Industrial Systems The efficient operation and extended lifespan of industrial machinery and equipment are intricately linked to precise maintenance practices and lubrication methodologies.

202

S. Yildiz et al.

Bearings, among the essential components, play a pivotal role in facilitating the smooth operation of moving parts within machinery [2].

Fig. 2. The causes of bearing losses

Reduction in Friction and Wear: Bearings substantially alleviate the friction encountered by rotating and moving machinery components, thereby ensuring their efficient functioning. Adequate lubrication minimizes surface contact and friction between these components, consequently reducing wear and tear on bearings. Wear and tear are significant factors contributing to the premature deterioration of bearings. Lubrication, by preserving smooth and protected bearing surfaces, guarantees the longevity of bearings [3]. Control of Heat Generation: Friction within bearings generates heat, which, when not adequately lubricated, can accumulate within the bearing, resulting in overheating. Overheating compromises the properties of the bearing, alters its material structure, and detrimentally affects its performance. Appropriate lubrication effectively manages the heat generated by friction, preventing bearing overheating and extending its operational life [4]. Enhanced Load Carrying Capacity: Bearings shoulder the responsibility of carrying loads within machinery. Optimal lubrication augments a bearing’s load-carrying capacity. The presence of an oil film facilitates improved load distribution across the bearing surface, reducing direct contact between the load and the surface. This empowers the bearing to withstand higher loads and mitigates the risk of damage due to overloading [5]. Protection against Contaminants: Bearings are shielded against the adverse effects of environmental elements, such as dirt, dust, and foreign particles. Statistics reveal that 50% of bearing failures arise from inadequate lubrication, neglect, and the use of unsuitable lubricating oils. Figure 2 illustrates the factors contributing to bearing failures [6].

Cost-Effectiveness of an Automatic Lubrication System

203

1.4 Lubrication Film Analysis in Oil Lubrication The primary parameter for assessing the lubrication condition is the thickness of the lubricating film present between the load-bearing surfaces engaged in rolling and sliding contact. The lubricant film between the rolling contact surfaces can be accurately characterized using the principles of elastohydrodynamic (EHD) lubrication theory.

Fig. 3. Oil film values [1]

On the other hand, lubrication in sliding contact conditions, such as that occurring between the faces of tapered roller bearings and their corresponding lips, can be adequately described by hydrodynamic lubrication theory, given that the contact pressure in these sliding areas is lower than in the rolling contact regions. Figure 3 illustrates the values employed in the calculations of the oil film thickness [1]. 1.5 Automatic Lubrication Systems and SmartLub Improper lubrication practices have resulted in a decline in maintenance efficiency across various sectors. While manual lubrication remains necessary for numerous production machines and specific lubrication points, it often leads to product breakdowns due to excessive lubricant application and high energy consumption. Within industrial operations, manual lubrication is typically employed for maintenance purposes; however, this approach often results in significant lubricant wastage, ultimately hampering production efficiency. In terms of worker safety and production speed, the benefits provided by automatic lubrication systems far outweigh those of manual methods. Automatic lubrication systems ensure a consistent and appropriate quantity of lubricant is applied to the system using grease guns. The implementation of automatic lubrication systems effectively safeguards bearings from failure by supplying the machine with the precise amount of oil at the appropriate intervals. This not only prevents over-lubrication but also yields energy and material

204

S. Yildiz et al.

savings, as well as improved quality and efficiency. Moreover, automatic lubrication systems control the excess lubricant quantity, ensuring the final product is not compromised within the worker area. Lubricating grease finds application across diverse industries to mitigate wear and friction between moving parts. Its semi-solid nature allows it to act as a seal, effectively preventing leakage [7]. The fundamental distinction between automatic and manual lubrication lies in the application method, with manually applied lubricants typically following a mechanical lubrication chart. Automation offers numerous advantages over manual approaches, particularly in terms of more controlled and precise lubricant application. It is preferable to apply smaller quantities of grease at frequent intervals rather than applying a large amount at once. In manual lubrication, the challenge lies in maximizing the relubrication intervals by applying as much grease as possible without causing damage due to excessive grease accumulation [8]. Sarsha and Tamboli conducted a research study aimed at addressing lubrication challenges associated with the ATC (Automatic Tool Changer) mechanical arm, which is utilized for the production of complex, high-quality, and efficient products. CNCs (Computer Numerical Control) are employed to regulate the manufacturing process, and ATCs automate the tool change operations, ensuring proper tool holding in both tool holders of the mechanical arm. Lubrication plays a vital role in reducing friction between the surfaces involved in relative motion [9]. Currently, the feeding systems of CNC machine tools are lubricated using fixed stroke or periodic oil supply methods. However, such lubrication approaches prove inadequate for high loads, high-speed movements, and situations involving excessive finishing and low feed rates. This research study focuses on determining the ideal timing for lubrication in the feeding system. Sensors are utilized to assess parameters such as servomotor torque value, current, precision, and oil film thickness during the operation of the feeding system. Furthermore, the sensor readings are employed to validate and construct a lubrication characteristic model, utilizing a back-propagation neural model to estimate the ideal lubrication condition. The obtained data is then processed and fed back to the machine tool controller, enabling the lubrication system to become more intelligent. Experimental results demonstrated the effectiveness of this approach, particularly with regards to the feed rate [10]. Railway wheels are manufactured through casting or forging processes, which involve upsetting, forging, rolling, heat treatment, and machining. The quality of metal flow during upsetting and forming stages is influenced by the surface finish of the dies and the level of friction experienced between the dies and the workpiece. Insufficient die lubrication during forging operations leads to inadequate die surfaces and uneven metal flow, resulting in reduced die life and increased rejection rates due to dimensional discrepancies. To address this issue, an autonomous die lubrication system was developed and implemented at a forging wheel and axle plant to enhance die life and improve dimensional accuracy. The design of the lubrication system was based on die wear patterns, and it receives a signal from the press following the completion of upsetting, shaping, and piercing operations [11]. The novel automated lubrication control system is specifically designed for computer numerical control (CNC) machine tool guideways, introducing a new approach to

Cost-Effectiveness of an Automatic Lubrication System

205

machining technology by Sparham et al. [12]. Failure detection and correction within the lubrication system are achieved through the utilization of sensitive temperature sensors installed in the machine tool guideways. These sensors provide temperature signals that reflect the friction, wear, and heat generated by the machine tool, enabling the identification and rectification of lubrication system issues. Study by Gritsenko et al. [13] have revealed that up to 27% of internal combustion engine (ICE) failures under current operating conditions can be attributed to turbocharger problems. Heavy-duty operation, and the limited load-bearing capacity of lubricants and structural materials, the failure rate of turbochargers is significantly high. To address this reliability issue, the implementation of an independent lubrication system with a hydraulic accumulator is recommended. A test bench was developed to monitor and assess oil output parameters, focusing on variables such as turbocharger rotor speed, oil pressure in front of the turbocharger bearing with and without a hydraulic accumulator, oil consumption, and other relevant factors. Wheel loaders are commonly employed in mining operations to extract high-quality ore. Unplanned downtimes of these primary digging units lead to production delays and financial losses. However, the integration of automatic lubrication systems on these machines helps mitigate frictional wear on critical components and extends maintenance intervals, thus contributing to improved operational efficiency and reduced downtime [14].

Fig. 4. Automatic Lubrication System and SmartLub Smart Control Systems Working Principle

206

S. Yildiz et al.

In the pursuit of enhancing the efficiency of machinery and plants while simultaneously reducing spare parts and maintenance expenditures within the grain milling sector, the SmartLub automatic lubrication system has emerged as a viable solution. This meticulously designed system incorporates an intelligent control unit, thereby ensuring precise and efficient lubrication concerning factors such as the appropriate volume, timing, and lubrication points. The development of this conclusion is predicated on an amalgamation of insights culled from comprehensive literature reviews and insights gleaned from bearing catalogues. SmartLub, in essence, epitomizes a cutting-edge lubrication system meticulously engineered for seamless integration into grain mills, with the promise of delivering heightened performance and substantial cost savings. As depicted in Fig. 4, a visual representation of the automatic lubrication system elucidates its intricate components and operational intricacies. The system’s core functionality revolves around the orchestrated transfer of oils from the central lubrication unit to the lubrication tanks situated within the rollers, and this is executed in a meticulously timed and periodic manner. This synchronized approach ensures that the lubrication process unfolds at predetermined intervals, a pivotal aspect in its effectiveness within grain milling operations.

2 Material and Method 2.1 Economic Contribution of Automatic Lubrication Systems Efficient lubrication practices play a pivotal role in minimizing costs associated with maintenance, lubricants, and machine parts. While manual lubrication systems have persisted despite ergonomic challenges and attempts at improvement, recent years have witnessed significant progress in lubrication techniques through the widespread adoption of automatic lubrication systems across diverse industries. Insufficient lubrication has emerged as a primary factor contributing to premature failure of tribomechanical components, potentially leading to machinery damage [15]. To mitigate such risks, selecting an appropriate automatic lubrication system and determining optimal lubricant dosage and application methods for specific usage conditions are imperative. This study aims to demonstrate the economic viability of implementing automatic lubrication systems for tribomechanical components. In terms of cost-effectiveness, automatic lubrication systems outperform manual methods. This advantage stems from the elimination of labor costs and the reduction in errors resulting from human intervention. To evaluate economic feasibility, the basic payback method is employed [16] given in Eq. 1. Payback Time =

Initial Investment Annual Net Income

(1)

In facilities lacking automatic lubrication systems, roller mills are required to undergo a one-day shutdown every two months for lubrication purposes, resulting in a cumulative downtime of six days per year. These interruptions are primarily caused by issues such as dirt contamination, insufficient lubricant supply, and faulty lubrication practices. In addition to the economic losses associated with these stoppages, the implementation of

Cost-Effectiveness of an Automatic Lubrication System

207

SmartLub brings about improvements in machine efficiency through enhanced lubrication quality and removal of contaminants. This leads to a reduction in failure rates and increased bearing lifespan. Compared to manual lubrication, where bearing life typically spans five years, the utilization of SmartLub has the potential to extend bearing longevity up to 20 years. 2.2 Calculation of Oil Requirements An oil film thickness calculation was performed by referencing the parameters delineated in Fig. 3, utilizing the point contact model proposed by Dowson. Employing the pertinent formulas and methodologies outlined in this analysis, the oil film thickness was meticulously computed. Furthermore, the requisite volume of oil necessary for the lubrication of a total of 22 × 8 bearings was calculated within the context of this investigation. It’s worth noting that the bearing subjected to scrutiny and evaluation in these calculations is the SFK 22315 [1]. Used Equations are given below.   (2) hmin = 3, 63 × U 0,68 × G 0,49 × W −0,073 × 1 − e−0,68×k × Rr [m] where, U = n0 × v/E  × Rr

(3)

G = a × E   W = Q/ E  × Rr 2

(4) (5)

  The modulus of elasticity was taken as 2, 08 × 1011 N/m2 for steel and Poisson’s ratio was taken 0,3 for steel. The lubrication system requirements is also calculated and 325 Bar pressure pump and 800 bar pressure resistant elements are used. When lubrication is required, the system is activated with sensors or a time plan, and automatic lubrication is provided. The selection of the oil pump was based on two crucial factors: the viscosity of the oil and the head loss within the piping system. Additionally, consideration was given to the head loss occurring as the oil traveled through the system’s components.

3 Results Discussion In the hypothetical scenario where the losses attributed to stoppages are excluded in a system with a daily capacity of 450 tons, encompassing 22 roller mills, each equipped with 8 bearings, conventional manual lubrication practices would typically yield an average bearing lifespan of 2 years. However, the application of the SmartLub system extends this lifespan to a minimum of 4 years, underscoring the profound significance of automatic lubrication, especially within manufacturing facilities where manual lubrication poses inherent challenges. In a configuration featuring 22 rolls, each housing 8 bearings valued at $225 per bearing, the adoption of SmartLub yields a substantial return of $39,600 every 4 years. This adoption also translates to labor cost savings amounting to $8,800 over the same interval. Considering the system’s average operational lifespan of 20 years, the cumulative savings reach an impressive sum of $242,000.

208

S. Yildiz et al. Table 1. System inputs for economic analysis

Total ready-to-use installation cost

15000 $

Annual savings

12100 $

Lifetime

20 years

Cost of a bearing

225 $

Salvage value after lifetime

–

Consequently, the annual profit is calculated at $12,100. Employing the simple payback formula, which accounts for the estimated $15,000 investment required for the SmartLub smart lubrication system, reveals an expeditious recouping of this expenditure within a mere 1.23 years. Notably, the scrap cost of the system remains unaccounted for in this analysis, as it is deemed negligible in its impact. The economic analysis specifications, along with annual cash flow projections spanning the entirety of the system’s operational lifespan, are presented in Table 1 and Table 2, respectively. Table 2. Annual cash flow in the life time of SmartLub application Items/Years

0

2

6-10-14

18

SmartLub Investment Cost

−15000 $

Bearing cost savings

39600 $

39600 $

39600 $

Labor Savings

8800 $

8800 $

8800 $

48400 $

48400 $

48400 $

Net Budget

−15000 $

4 Conclusion In conclusion, this study has elucidated the pivotal role of lubrication in enhancing the performance and economic efficiency of grain milling systems. Through the implementation of the SmartLub automatic lubrication system, remarkable improvements in both bearing lifespan and operational costs have been demonstrated. In a system with 22 roller mills, each comprising 8 bearings, the adoption of SmartLub has extended the average bearing life from 2 to 4 years, resulting in a substantial return on investment. Over a 4-year period, this extended bearing lifespan translates to a significant savings of $39,600, while labor costs are concurrently reduced by $8,800. Over the system’s average operational lifespan of 20 years, the cumulative savings amount to an impressive $242,000, with an annual profit of $12,100. The quick return on investment is evidenced by a payback period of just 1.23 years for the lubrication system, which incurs an estimated cost of $15,000. Moreover, the economic viability of the system is further underscored by the minimal consideration of scrap costs, as they are deemed negligible in the overall cost-benefit

Cost-Effectiveness of an Automatic Lubrication System

209

analysis. The economic analysis specifications, alongside annual cash flow projections spanning the system’s operational lifespan, have been meticulously presented. In light of these findings, it is evident that the adoption of advanced lubrication systems such as SmartLub holds immense potential for revolutionizing the grain milling sector. Beyond extending bearing lifespans and reducing maintenance costs, automatic lubrication contributes to enhanced machine efficiency, ensuring optimal lubrication even in challenging operational environments. As such, the adoption of automated lubrication technologies emerges as a compelling strategy for industries reliant on highperformance machinery, promising not only cost savings but also improved reliability and profitability.

References 1. FAG, Rolling Bearing Lubrication, Publ. No. WL 81 115/4 EA Edition (2002) 2. Smith, H.P.: Farm Machinery and Equipment. Read Books Ltd. (2020) 3. Booker, J.D., Mellor, P.H., Wrobel, R., Drury, D.: A compact, high efficiency contra-rotating generator suitable for wind turbines in the urban environment. Renew. Energy 35(9), 2027– 2033 (2010) 4. Liu, Z., Zhang, L.: A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings. Measurement 149, 107002 (2020) 5. Litwin, W.: Influence of local bush wear on water lubricated sliding bearing load carrying capacity. Tribol. Int. 103, 352–358 (2016) 6. SKF - PUB MP/P1 03000 ES - Bearing Maintenance (2022) 7. Mewis, J., Wagner, N.J.: Thixotropy. Adv. Coll. Interface. Sci. 147, 214–227 (2009) 8. Wakiru, J.M., Pintelon, L., Muchiri, P.N., Chemweno, P.K.: A review on lubricant condition monitoring information analysis for maintenance decision support. Mech. Syst. Signal Process. 118, 108–132 (2019) 9. Sardhara, T., Tamboli, K.: Design and development of automatic lubrication system for ATC of CNC. Mater. Today: Proc. 5(2), 3959–3964 (2018) 10. Chen, S.H., Haung, Z.J.: A study of using back-propagation neural model in automatic lubrication installation for the feeding system of computer numerical control machine tool. Proc. Inst. Mech. Eng. Part J: J. Eng. Tribol. 236(6), 1219–1231 (2022) 11. Jha, S.K., Jain, D.K., De, S., Chakraborty, M., Karmakar, D.: Modified die lubrication system for forging wheel and axle plant. In: Prasad, R., Sahu, R., Sahoo, K. L., Jadhav, G. N. (eds.) Advancement in Materials Processing Technology. SPM, vol. 12, pp. 205–211. Springer, Singapore (2022). https://doi.org/10.1007/978-981-16-3297-6_20 12. Sparham, M., Sarhan, A.A., Mardi, N.A., Hamdi, M.: Designing and manufacturing an automated lubrication control system in CNC machine tool guideways for more precise machining and less oil consumption. Int. J. Adv. Manuf. Technol. 70, 1081–1090 (2014) 13. Gritsenko, A., Almetova, Z., Burzev, A., Tsybunov, E.: Investigation of the independent lubrication system of modern turbocharged vehicles. Transp. Res. Procedia 57, 250–255 (2021) 14. Warner, C., Desmet, A.: Automated lubrication systems prognostics using long-term recurrent convolutional networks. In: IEEE International Conference on Prognostics and Health Management (ICPHM), pp. 1–8. IEEE (2018) 15. Ašonja, M.S.A., Adamovi´c, P.D.Ž.: The economic justification of the automatic lubrication using. In: 14th International Research/Expert Conference” Trends in the Development of Machinery and Associated Technology” TMT 2010, Mediterranean Cruise, pp. 11–18 (2010) 16. Aksoy, M.H., ˙Ispir, M.: Techno-economic feasibility of different photovoltaic technologies. Appl. Eng. Lett. 8(1), 1–9 (2023)

Combining DOE and EDAS Methods for Multi-criteria Decision Making Do Duc Trung1

, Nguyen Xuan Truong1 , Hoang Tien Dung1 and Aleksandar Ašonja2(B)

,

1 Faculty of Mechanical Engineering, Hanoi University of Industry, Hanoi 100000, Vietnam 2 Faculty of Economics and Engineering Management in Novi Sad, University Business

Academy in Novi Sad, Novi Sad, Republic of Serbia [email protected]

Abstract. Most existing Multi-Criteria Decision Making (MCDM) methods can only be used to rank readily available choices. After the first ranking has been completed, some of the options are discarded and others may arise, then it will be necessary to start the ranking over again. This study proposes a new method that can quickly rank alternatives in this situation. The proposed method is a combination of the Design of the Experimental method (DOE) and Evaluation based on the Distance from Average Solution (EDAS) method, hence it will be referred to as the DOE-EDAS method. The new aspect of this proposed method is to build a relationship between the scores of the options and the criteria. This relationship is called the regression equation. Using this regression equation, it is possible to quickly calculate the scores for alternatives (after an alternative is discarded or an alternative is added) and rank them. The method has been applied to multi-criteria decision-making in some specific cases. The results demonstrate the effectiveness of this method. Keywords: MCDM · EDAS method · DOE-EDAS method · Multi-Criteria Decision Making

1 Introduction When there are many options for evaluating a product or process, deciding on the optimal option is a demand in all subject domains. The job will be simple if each option uses only one evaluation criterion. However, in reality, each choice often has different criteria that complicate the evaluation. Over the years, hundreds of MCDM methods have been developed to solve problems in different situations [1]. While popular methods like VIKOR or TOPSIS usually calculate the distance from the alternatives to the ideal solution or the opposite of the ideal method, the EDAS method measures the distance from the other options to the mean; thus, it is considered to be more beneficial than the above two methods for decision making when there are a lot of conflicts between the alternatives [2, 3]. Many studies have applied this method to evaluate/rank options in different fields, for example, supplier selection [4, 5], selection of subcontractors in construction activities [6], selection of locations for hydrogen © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 210–227, 2024. https://doi.org/10.1007/978-3-031-51494-4_19

Combining DOE and EDAS Methods

211

refueling stations for electric vehicle traffic [7], ranking of solid waste management systems in Nordic countries [8], evaluation of energy-saving green building design projects [9], ranking of milling processes in mechanical processing [10], determination of the percentage of additives mixed into the compound composite materials [11], choosing a transportation development strategy to reduce carbon emissions to the environment [12], etc. In some of the studies mentioned above, the EDAS has successfully ranked alternatives in many fields. However, like all other MCDM methods, the studies applying the EDAS method have only been used to rank the available alternatives. To be more specific, the current MCDM methods (including EDAS method) only consider the ranking of the available alternatives and do not consider the case when one/several alternatives are rejected, approved, or added to the list. After the ranking of options is completed and new options are added, the ranking must start over again. This work must also be repeated if one/several options are removed from the list. This is seen as a general limitation of existing MCDM methods. Given this fact, this study proposes a solution to overcome this problem. The proposed method combines DOE and EDAS, hence named the DOEEDAS method. The second part of this paper presents the steps to rank alternatives by the EDAS method. The details of the proposed method (DOE-EDAS) are discussed in the third part. The effectiveness of the proposed method is assessed in Sect. 4 through some specific examples. Finally, the fifth part of this paper presents the conclusions of this study and some directions for future research.

2 EDAS Method The EDAS method ranks the alternatives in the following order [13]: Step 1: The decision matrix is built as in Table 1. Table 1. Decision Matrix. No.

C1

C2

Cj

Cn

A1

x11

x12

x1j

x1n

A2

x21

x22

x2j

x2n

Ai

xi1

xi2

xij

xin

Am

xm1

xm2

xmj

xmn

Where: m, n are the number of options and the number of criteria, respectively; x ij is the value of the criterion C j in the Ai option. Step 2: Calculate the average value for each criterion according to formula (1). m xi AVG = i=1 (1) m Step 3: Calculate the positive distance (PD) and negative distance (ND) from the mean.

212

D. D. Trung et al.

– If j is the criterion, the bigger the better.    max 0, xij − AVG j PDij = AVG j    max 0, AVG j − xij NDij = AVG j – If j is the criterion, the smaller the better.    max 0, AVG j − xij PDij = AVG j    max 0, xij − AVG j NDij = AVG j

(2) (3)

(4) (5)

Step 4: Calculate the total positive distance (SoP) and total negative distance (SoN) according to two formulas (6) and (7). m SoP i = wj · PDij (6) i=1

SoN i =

m i=1

wj · NDij

(7)

where wj is the weight of the jth indicator. Step 5: Calculate the quantities SoP and SoN according to two corresponding formulas (8) and (9). SSoP i =

SoP i max(SoP i )

SSoN i = 1 −

SoN i max(SoN i )

(8) (9)

Step 6: The score of each alternative (APS) is calculated using this formula (10). APS i =

1 (SSoP i + SSoN i ) 2

(10)

Step 7: Rank the alternatives according to the principle such that the one with the largest is the best one.

Combining DOE and EDAS Methods

213

3 Recommended Method Step 1: Build the decision matrix as shown in Table 1. Step 2: Determine the low and high for each criterion as shown in Table 2. Table 2. Levels of parameters. No.

Value at levels Low

Hight

C1

Min(C i1 ), i = 1 ÷ m

Max(C i1 ), i = 1 ÷ m

C2

Min(C i2 ), i = 1 ÷ m

Max(C i2 ), i = 1 ÷ m

Cj

Min(C ij ), i = 1 ÷ m

Max(C ij ), i = 1 ÷ m

Cn

Min(C in ), i = 1 ÷ m

Max(C in ), i = 1 ÷ m

Step 3: Build an experiment matrix with input parameters as the criteria. The maximum/minimum value of each input parameter and its corresponding value is determined in step 2 (Table 2). Step 4: Calculate scores from all experiments. This step is the combined procedure from step 2 to step 6 of the EDAS method (presented in part 2). Step 5: Build the relationship between the scores of the experiments with the criteria (11).   (11) APS i = f Cj , j = 1 ÷ n Step 6: Use formula (11) to calculate the score for the options in Table 1, known as the DOE-EDAS score. Step 7. Rank the alternatives with DOE-EDAS score. The best solution is the one with the highest score, and vice versa. Formula (11) calculates the score for the available options. However, as an option is added/removed from the list of options, this formula is also used to calculate the score for the remaining alternatives in the list of alternatives. This is an innovative approach that this study has discovered. Some examples are presented in the next part of this paper to evaluate the proposed method’s effectiveness.

4 Applied Cases 4.1 Case 1 In this case, fourteen options for office climate ratings were used [14]. Six criteria were used to evaluate each option: The amount of air per head (TH), relative air humidity (RH), air temperature (AT), illumination during work hours (IH), rate of airflow (RF), and dew points (DP). The first four criteria are the “the-greater-the-better”, and the last two are the “the-smaller-the-better”. The ranking of alternatives was also done by CODAS method, with the weights of the criteria chosen randomly, using the same value as in the last row of Table 3 [14].

214

D. D. Trung et al. Table 3. Data of Case 1 [14].

No.

TH

RH

AT

IH

RF

DP

A1

7.6

46

18

390

0.1

11

A2

5.5

32

21

360

0.05

11

A3

5.3

32

21

290

0.05

11

A4

5.7

37

19

270

0.05

9

A5

4.2

38

19

240

0.1

8

A6

4.4

38

19

260

0.1

8

A7

3.9

42

16

270

0.1

5

A8

7.9

44

20

400

0.05

6

A9

8.1

44

20

380

0.05

6

A10

4.5

46

18

320

0.1

7

A11

5.7

48

20

320

0.05

11

A12

5.2

48

20

310

0.05

11

A13

7.1

49

19

280

0.1

12

A14

6.9

50

16

250

0.05

10

Random Weight

0.21

0.16

0.26

0.17

0.12

0.08

The ranking of alternatives by the DOE-EDAS method is done as follows: – Determine the low and high levels of each criterion. The results are detailed in Table 4. Table 4. Highest and lowest values for each criterion. No.

Value at levels Low (−1)

Hight (+1)

TH

3.9

8.1

RH

32

50

AT

16

21

IH

240

400

RE

0.05

0.1

DP

5

12

– Build an experimental matrix with the main input parameters being six criteria for each option. The entire two-level experiment plan (2k ) was used to construct the experimental matrix. This is the simplest form of planning but still ensures the accuracy of building the relationship between output parameters (scores of alternatives) and input parameters (criteria) during the experiment [15]. The two levels here are

Combining DOE and EDAS Methods

215

understood as the lowest level (corresponding to encryption level −1) and the highest level (corresponding to encoding level +1). In this case, the number of criteria equals six, so the experimental matrix will consist of sixty-four experiments. Part of the experimental matrix is summarized in Table 5. Table 5. Part of the experimental matrix and EDAS scores of each experiment. Exp.

TH

RH

AT

IH

RF

DP

APSi

1

3.9

32

21

240

0.1

5

0.2626

2

8.1

32

21

240

0.05

12

0.5734

3

3.9

32

21

400

0.1

12

0.2995

4

3.9

32

16

400

0.05

5

0.4454

5

3.9

50

16

400

0.1

5

0.4266

6

3.9

50

21

240

0.05

5

0.5525

7

8.1

32

21

240

0.05

5

0.7005

8

8.1

32

21

400

0.1

5

0.7102

…

…

…

…

…

…

…

…

63

3.9

32

16

400

0.05

12

0.3183

64

3.9

50

16

240

0.1

12

0.1355

– Scores of each experiment (APSi ) were calculated following the EDAS method’s steps, and is also summarized in Table 5. Table 5 illustrates the relationship between APSi score and the criteria by formula (12). To evaluate the accuracy of Eq. (12), it is necessary to base it on three parameters: R-Sq, R-Sq(pred), and R-Sq(adj). The values of these three parameters are always greater than zero. Equations are said to have a higher level of accuracy as their values get closer to 1 [15]. In this case, all three parameters have a value of 1, which means that Eq. (12) has a very high precision. APS i = − 0.65741 + 0.06751 · TH + 0.00752 · RH + 0.02711 · AT + 0.00102 · IH − 3.08644 · RF − 0.01815 · DP

(12)

Formula (12) calculates DOE-EDAS scores for fourteen options in Table 3. The results are presented in Table 6. Also, the results of ranking options by DOE-EDAS method have been obtained, performed, and are summarized in Table 6.

216

D. D. Trung et al. Table 6. Ranking of alternatives of case 1 by DOE-EDAS method.

No.

TH

RH

AT

IH

RF

DP

DOE-EDAS APS

Rank

A1

7.6

46

18

390

0.1

11

0.57907

4

A2

5.5

32

21

360

0.05

11

0.53707

7

A3

5.3

32

21

290

0.05

11

0.45217

10

A4

5.7

37

19

270

0.05

9

0.47846

8

A5

4.2

38

19

240

0.1

8

0.21794

14

A6

4.4

38

19

260

0.1

8

0.25184

12

A7

3.9

42

16

270

0.1

5

0.23149

13

A8

7.9

44

20

400

0.05

6

0.89378

1

A9

8.1

44

20

380

0.05

6

0.88688

2

A10

4.5

46

18

320

0.1

7

0.37099

11

A11

5.7

48

20

320

0.05

11

0.60299

3

A12

5.2

48

20

310

0.05

11

0.55903

5

A13

7.1

49

19

280

0.1

12

0.46464

9

A14

6.9

50

16

250

0.05

10

0.53735

6

The ranking of alternatives by EDAS method has also been carried out. Table 7 presents the ranking of the alternatives by several different methods. According to Table 7: – 12/14 alternatives are ranked the same when comparing the DOE-EDAS and EDAS methods (except A2 and A14 ). – The ratings 1 (A8 ), 2 (A9 ), and worst (A5 ) options are the same when using three different methods. Thus, determining the best alternative using the proposed method (DOE-EDAS) was successful in this case. However, for a comprehensive assessment of the effectiveness of a proposed method, sensitivity analysis should be performed [16, 17]. Changing the weight of the criteria was chosen for the sensitivity analysis in this case [18, 19]. In addition to the set of weights used above, three other sets of weights determined by three different methods were also used, including the Equal weighting method, the Entropy weighting method, and the MEREC weighting method. Equal weighting is the simplest method of determining the weights with equal criteria weights. There are six criteria in this case, so all criteria weigh 1/6. Two methods, Entropy weighting and MEREC weighting were used because they are the recommended methods [20]. The weighting steps in these methods can be found in many research papers [21, 22]. Table 8 presents the weights of the criteria that different methods have determined. Using the three sets of weights in Table 8, three equations were built showing the relationship between the DOE-EDAS scores of the options with the following criteria:

Combining DOE and EDAS Methods

217

Table 7. Ranking of the alternatives of case 1. No.

DOE-EDAS

EDAS

CODAS [14]

A1

4

4

3

A2

7

6

7

A3

10

10

9

A4

8

8

10

A5

14

14

14

A6

12

12

13

A7

13

13

12

A8

1

1

1

A9

2

2

2

A10

11

11

11

A11

3

3

4

A12

5

5

6

A13

9

9

8

A14

6

7

5

Table 8. The weights of the criteria are determined by different methods. Weight method

TH

RH

AT

IH

RF

DP

Equal

1/6

1/6

1/6

1/6

1/6

1/6

Entropy

0.0429

0.2324

0.1843

0.2719

0.1570

0.1115

MEREC

0.1773

0.1292

0.0794

0.1118

0.3467

0.1556

– When using Equal weights. APS i = − 0.08832 + 0.04902 · TH + 0.00717 · RH + 0.01590 · AT + 0.00091 · IH − 3.92216 · RF − 0.003460 · DP

(13)

– When using Entropy weights. APS i = − 0.40023 + 0.01390 · TH + 0.01102 · RH + 0.01936 · AT + 0.00165 · IH − 4.07018 · RF − 0.02550 · DP

(14)

– When using MEREC weights APS i = 0.50745 + 0.04785 · TH + 0.00510 · RH + 0.00695 · AT + 0.00056 · IH − 7.48651 · RF − 0.02964 · DP

(15)

218

D. D. Trung et al.

Again, these three formulas are used to calculate the DOE-EDAS scores for the alternatives, making it possible to rank them, as shown in Table 9. For the convenience of discussion, the evaluation results when using randomly selected weights (calculated above, in Table 7) have also been aggregated in this table. In addition, the ranking results when using different sets of weights have also been presented in the graph in Fig. 1. The data in Table 9 and Fig. 1 show that the ranking results of the alternatives change very little for different sets of weights. Particularly, the best option (A8 ), option 2 (A9 ) and the worst one (A5 ) are always determined uniformly with four different sets of weights. Thus, it can be concluded that the proposed method was successful in this case. Table 9. Ranking of alternatives of case 1 by DEO-EDAS method with different weight sets. No.

Random

Equal

Entropy

MEREC

A1

4

8

6

9

A2

7

7

5

7

A3

10

9

10

8

A4

8

6

8

6

A5

14

14

14

14

A6

12

13

13

13

A7

13

12

12

12

A8

1

1

1

1

A9

2

2

2

2

A10

11

10

9

10

A11

3

3

3

4

A12

5

5

4

5

A13

9

11

11

11

A14

6

4

7

3

4.2 Case 2 In this case, the data from nine variants of the turning process were used, as shown in Table 10 [23]. Each alternative is described with four criteria consisting of three shear force components in the three directions x, y, and z are Fx, Fy, and Fz, respectively, together with the material removal rate MRR. The first three criteria belong to the thesmaller-the-better category, whereas the last one is of the-greater-the-better. The ranking of the alternatives is to find the solution that simultaneously ensures the first three criteria are considered the smallest and the last criterion is considered the largest. Seven other methods, including SAW, WASPAS, TOPSIS, VIKOR, MOORA, COPRAS, and PIV, were also used to perform this task, with the weights of the criteria determined by the Entropy method having values as given in the last row of Table 10 [23].

Combining DOE and EDAS Methods

219

Fig. 1. Ranking of the alternatives of case 1 by DEO-EDAS method with different weight sets.

Table 10. Data of Case 2 [23]. No.

Fx

Fy

Fz

A1

59.844

187.437

44.165

A2

87.943

199.762

99.125

49.062

A3

78.913

127.456

69.874

109.108

A4

54.816

172.714

60.19

28.588

A5

63.117

180.361

68.869

99.039

A6

68.79

113.951

70.694

61.669

A7

46.654

116.88

92.222

57.177

A8

44.989

162.337

63.25

55.462

A9

54.846

167.837

74.165

Entropy weight

0.2427

0.2598

0.2462

MRR 11.561

151.09 0.2514

Doing the same as with case 1, an equation showing the relationship between scores of experiments with four criteria can be constructed (16). APS i =0.62573 − 0.00250 · Fx − 0.00113 · Fz − 0.00235 · Fz + 0.00476 · MRR

(16)

Using (16) to calculate the scores for nine vehicle alternatives, the results are presented in Table 11. The ranking results of the alternatives are also summarized in this table. Table 12: results of ranking the alternatives of case 2 by several different methods. The results in Table 12 show that: – The worst-case option (A2 ) determined by the DOE-EDAS method is the same as that determined by the SAW, WASPAS, TOPSIS, VIKOR, and PIV methods.

220

D. D. Trung et al. Table 11. Score of alternatives and ratings.

No.

Fx

Fy

Fz

MRR

APS

Rank

A1

59.844

187.437

44.165

11.561

0.21556

8

A2

87.943

199.762

99.125

49.062

0.18073

9

A3

78.913

127.456

69.874

109.108

0.63957

2

A4

54.816

172.714

60.19

28.588

0.28816

7

A5

63.117

180.361

68.869

99.039

0.57371

3

A6

68.79

113.951

70.694

61.669

0.45240

4

A7

46.654

116.88

92.222

57.177

0.43246

6

A8

44.989

162.337

63.25

55.462

0.44518

5

A9

54.846

167.837

74.165

151.09

0.84386

1

Table 12. Ranking of the alternatives of case 2. No. DOE-EDAS EDAS Rank [23] SAW WASPAS TOPSIS VIKOR MOORA COPRAS PIV A1

8

9

7

8

8

8

9

9

8

A2

9

8

9

9

9

9

8

7

9

A3

2

2

2

2

2

5

2

2

2

A4

7

7

8

7

7

6

7

8

7

A5

3

3

6

6

3

7

3

4

3

A6

4

6

5

5

4

3

4

3

5

A7

6

4

3

3

5

4

5

5

6

A8

5

5

4

4

6

2

6

6

4

A9

1

1

1

1

1

1

1

1

1

– The ranking option 8 (A1 ) results from DOE-EDAS, WASPAS, TOPSIS, VIKOR, and PIV methods. – Especially, all nine methods have identified A9 as the best option. A3 has also been identified as a rank 2 alternative using eight different methods, except for the VIKOR method. Thus, it can be affirmed that using the DOE-EDAS method has achieved the goal of determining the best solution in this case. Re-sensitivity analysis was performed using different weighting methods. In addition to the previous Entropy weighting method, two other methods will be used again: the Equal weighting and the MEREC weighting. Table 13 presents the weighted values of the criteria according to two different methods.

Combining DOE and EDAS Methods

221

Table 13. Weight of the criteria of case 2. Weight method

Fx

Fy

Fz

MRR

Equal

1/4

1/4

1/4

1/4

MEREC

0.0707

0.0375

0.0861

0.8057

Corresponding to the two sets of weights of the criteria in Table 13, two equations to calculate the scores for the alternatives were built as follows: – When using the Equal weight set. APS i = 0.62823 − 0.00258 · Fx − 0.00109 · Fz − 0.00239 · Fz + 0.00475 · MRR

(17)

– When using the MEREC weight set. APS i = 0.02050 − 4.05115 · 10−4 · Fx − 9.10514 · 10−5 · Fz − 4.57694 · 10−4 · Fz + 0.00680 · MRR

(18)

Equations (17) and (18) are again used to calculate the EDAS scores for the nine vehicle alternatives. From that score, the alternatives corresponding to these two sets of weights were also ranked, as shown in Table 14. Table 14. Ranking of alternatives of case 2 by DEO-EDAS method with different weight sets. No.

Equal

MEREC

Entropy

A1

8

9

8

A2

9

7

9

A3

2

2

2

A4

7

8

7

A5

3

3

3

A6

4

4

4

A7

6

5

6

A8

5

6

5

A9

1

1

1

The results of ranking the alternatives using the set of entropy weights (calculated above) were also summarised in Table 14. For the convenience of observation, the ranking results of the alternatives when using three different sets of weights have also been shown in Fig. 2. According to the data in Table 14 and Fig. 2:

222

D. D. Trung et al.

Fig. 2. Ranking of the alternatives of case 2 by DEO-EDAS method with different weight sets.

– The ranking results of the alternatives are identical when using two weighting methods, Equal weight and Entropy weight. – When using MEREC weights, the options of rank 1 (A9 ), rank 2 (A3 ), rank 3 (A5 ) and rank 4 (A6 ) are the same as when using the other two sets of weights. Thus, it can be concluded with certainty that the proposed method succeeded in this case. Through the two examples above, we have two important conclusions: – The best alternative determined using the DOE-EDAS method is always similar to other methods. – When using the DOE-EDAS method, the ranking results of the alternatives differ very little for different sets of weights. We have enough bases from the two cases above to conclude that the proposed method is reliable for multi-criteria decision-making. However, there are more outstanding advantages of the proposed method. The outstanding advantage of the proposed method is that its application will be much simpler than the EDAS method when one/or a few options are added/removed from the list of options. A case conducted shortly afterward will make this advantage more clear. 4.3 Case 3 In this case, data on nine variants of the grinding process were used [24], as shown in Table 15. Where Ra is the surface texture, Ax, Ay and Az are the vibrations of the grinding machine spindle, corresponding to the three directions x, y, and z, respectively. All four criteria belong to the smaller-the-better category. In contrast, MRR is the material removal capacity, which is of the greater-the-better criterion. The weights of the selected criteria are similar, i.e. 0.2. The TOPSIS method has also been used for multi-criteria decision-making [24]. In this case, multi-criteria decision-making is also conducted using the DOE-EDAS

Combining DOE and EDAS Methods

223

Table 15. Data of Case 3 [24]. No.

Ra

Ax

Ay

Az

MRR

A1

0.569

0.939

8.424

0.954

36.9236

A2

0.681

0.890

3.621

0.941

147.6944

A3

0.876

0.945

9.059

3.201

332.3125

A4

0.373

0.581

2.975

1.541

100.4250

A5

0.644

0.980

2.322

1.871

66.9500

A6

0.305

0.477

1.773

0.709

200.8500

A7

1.127

3.026

2.221

3.862

62.3519

A8

1.190

1.244

4.151

2.262

187.0556

A9

1.445

0.792

2.055

1.465

93.5278

Weight

0.2

0.2

0.2

0.2

0.2

method. To evaluate the advantages of the DOE-EDAS method, we assume that only eight options are offered for ranking in the initial stage. Thus, we will exclude one of the nine options from Table 15. Randomly select an alternative to discard, say A5. Then there are eight options left, as shown in Table 16. Table 16. Data of alternatives after removing A5 [24]. No.

Ra

Ax

Ay

Az

MRR

A1

0.569

0.939

8.424

0.954

36.9236

A2

0.681

0.890

3.621

0.941

147.6944

A3

0.876

0.945

9.059

3.201

332.3125

A4

0.373

0.581

2.975

1.541

100.4250

A6

0.305

0.477

1.773

0.709

200.8500

A7

1.127

3.026

2.221

3.862

62.3519

A8

1.190

1.244

4.151

2.262

187.0556

A9

1.445

0.792

2.055

1.465

93.5278

Weight

0.2

0.2

0.2

0.2

0.2

Performing the same steps as the two cases above, i.e. building the experiment matrix, calculating the score for each experiment and building the regression equation, Eq. (19) is derived. All three parameters of this equation, including R-Sq, R-Sq(pred) and R-Sq(adj) are equal to 1, which shows that this equation has very high accuracy. APS i = 0.92384 − 0.16146 · Ra − 0.08066 · Ax − 0.02608 · Ay − 0.06230 · Az + 0.00076 · MRR

(19)

224

D. D. Trung et al.

After the Eq. (19) has been constructed, the previously eliminated alternative (A5 ) is added to the list. Using Eq. (19) to calculate the scores for all nine options and the results are presented in Table 17. Since then, the results from ranking the options by the DOEEDAS method have also been determined and summarized in this table. The results of ranking options for the EDAS and TOPSIS methods [24] have also been compiled into this table. Table 17. Ranking of the alternatives of case 3. No.

DOE-EDAS

EDAS

TOPSIS [24]

APSi

Rank

A1

0.5052

8

8

8

A2

0.7013

3

3

2

A3

0.5231

7

6

6

A4

0.7195

2

2

3

A5

0.6146

4

4

4

A6

0.8984

1

1

1

A7

0.2467

9

9

9

A8

0.5243

6

7

7

A9

0.5529

5

5

5

As the data in Table 17 shows, options A1 , A5 , A6 , A7 and A9 have identical rankings when using three different methods. Most importantly, all three methods identified A6 as the best option. Thus, it can be said that the multi-criteria decision-making task has been completed. In other words, the DOE-EDAS method was also successful in this case. This conclusion has demonstrated an outstanding advantage of the DOE-EDAS method that all existing MCDM methods have been unable to achieve. So, it is possible to quickly rank the alternatives when there one/several alternatives are added to the list of options. After evaluating the effectiveness of the DOE-EDAS method through the three cases above, it is shown that: Firstly, the best alternative determined by the DOE-EDAS method is always similar to the one produced by other MCDM methods. This demonstrates the accuracy of the proposed approach. Secondly, rank inversion occurs rarely (as in cases 1 and 2), and it is important that even though it is performed in different scenarios, all of them can yield the same best plan. Thirdly, the outstanding advantage of the proposed method is that it is possible to quickly rank the alternatives when there is a new addition/ change (in case 3). The ranking of alternatives as they are added to the list based on the regression equation results in a much faster decision-making speed than existing MCDM methods.

Combining DOE and EDAS Methods

225

5 Conclusion Choosing the best option among alternatives is a demand in all fields. Many MCDM methods have been developed to accomplish this task. However, until now the published studies only consider the case when the number of options to be ranked must be known before making a decision. In other words, after ranking the options, and one/several options are added, the decision-making needs to be recalculated from the beginning. Then, the decision-making process will be lengthy. This is an inconvenience in cases where deciding a short time/urgency is necessary. This paper proposes a new approach to overcome this limitation. First, the DOE method was used to build the experimental matrix with the input parameters as the criteria of the options. Next, the EDAS method was used to calculate the scores for each experiment. The third stage is to build the relationship between the scores of the alternatives and the criteria. This relationship is used to recalculate the scores of the alternatives, and thus the ranking of the alternatives is determined. When one/several options are added to the list of options, simply use the regression equation to calculate the scores for the alternatives, and the ranking of the options is done quickly. The DOEEDAS method has been evaluated for effectiveness by applying it in three different cases. Some conclusions are drawn as follows: – The best alternative determined by the DOE-EDAS method is always similar to other MCDM methods. – Very few rank reversals occur when using the DOE-EDAS method. Furthermore, the important thing is that the best solution is always determined uniformly in different scenarios. – The quick ranking when one/or a few options are added to the list of options is the noteworthy advantage of the proposed method. – Another research direction is also expected to be successful when combining DOE with another MCDM method and needs to be done shortly. – The biggest limitation of this study is that when an alternative is added, the minimum/maximum value of criterion j of that alternative is less/greater than all the minimum/maximum values of criterion j of the previous alternatives. Then, using the regression equation to calculate the scores for the additional solutions takes time to ensure accuracy. This limitation should be overcome as soon as possible.

References 1. Zopounidis, C., Doumpos, M.: Multiple Criteria Decision Making - Applications in Management and Engineering. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-392 92-9 2. Li, X., Ju, Y., Ju, D., Zhang, W., Dong, P., Wang, A.: Multi-attribute group decision making method based on EDAS under picture fuzzy environment. IEEE Access 7, 141179–141192 (2019). https://doi.org/10.1109/ACCESS.2019.2943348 3. Keshavarz Ghorabaee, M., Amiri, M., Zavadskas, E.K., Turskis, Z., Antucheviciene, J.: Stochastic EDAS method for multi-criteria decision-making with normally distributed data. J. Intell. Fuzzy Syst. 33(3), 1627–1638 (2017). https://doi.org/10.3233/JIFS-17184

226

D. D. Trung et al.

4. Xu, D., Cui, X., Xian, H.: An extended EDAS method with a single-valued complex neutrosophic set and its application in green supplier selection. Mathematics 8, 282 (2020). https:// doi.org/10.3390/math8020282 5. Keshavarz Ghorabaee, M., Zavadskas, E.K., Amiri, M., Turskis, Z.: Extended EDAS method for fuzzy multi-criteria decision-making: an application to supplier selection. Int. J. Comput. Commun. Control 11(3), 358–371 (2016). https://doi.org/10.15837/ijccc.2016.3.2557 6. Keshavarz-Ghorabaee, M., Amiri, M., Zavadskas, E.K., Turskis, Z., Antucheviciene, J.: A dynamic fuzzy approach based on the EDAS method for multi-criteria subcontractor evaluation. Information 9, 68 (2018). https://doi.org/10.3390/info9030068 7. Schitea, D., Deveci, M., Iordache, M., Bilgil, K., Akyurt, I.Z., Iordache, I.: Hydrogen mobility roll-up site selection using intuitionistic fuzzy sets based WASPAS, COPRAS and EDAS. Int. J. Hydrogen Energy 44(16), 8585–8600 (2019). https://doi.org/10.1016/j.ijhydene.2019. 02.011 8. Behzad, M., Zolfani, S.H., Pamucar, D., Behzad, M.: A comparative assessment of solid waste performance management in the nordic countries based on BWM-EDAS. J. Clean. Prod. 266, 122008 (2020). https://doi.org/10.1016/j.jclepro.2020.122008 9. Liang, Y.: An EDAS method for multiple attribute group decision-making under intuitionistic fuzzy environment and its application for evaluating green building energy-saving design projects. Symmetry 12, 484 (2020). https://doi.org/10.3390/sym12030484 10. Trung, D.D.: Application of EDAS, MARCOS, TOPSIS, MOORA and PIV Methods for multi-criteria decision making in milling process. Strojnícky cˇ asopis – J. Mech. Eng. 71(2), 69–84 (2021). https://doi.org/10.2478/scjme-2021-0019 11. Chairman, C.A., et al.: Mechanical and abrasive wear performance of titanium di-oxide filled woven glass fibre reinforced polymer composites by using taguchi and EDAS approach. Materials 14, 5257 (2021). https://doi.org/10.3390/ma14185257 12. Krishankumar, R., Pamucar, D., Deveci, M., Ravichandran, K.S.: Prioritization of zero-carbon measures for sustainable urban mobility using integrated double hierarchy decision framework and EDAS approach. Sci. Total Environ. 797(25), 149068 (2021). https://doi.org/10.1016/j. scitotenv.2021.149068 13. Ghorabaee, M.K., Zavadskas, E.K., Olfat, L., Turskis, Z.: Multi-criteria inventory classification using a new method of evaluation based on distance from average solution (EDAS). Informatica 26(3), 435–451 (2015). https://doi.org/10.15388/Informatica.2015.57 14. Keshavarz Ghorabaee, M., Zavadskas, E.K., Turskis, Z., Antucheviciene, J.: A new combinative distance-based assessment (CODAS) method for multi-criteria decision-making. Econom. Comput. Econom. Cybernet. Stud. Res. 50(3), 25–34 (2016) 15. Du, N.V., Binh, N.D.: Design of Experiment Techniques. Science and Technics Publishing House, Ha Noi (2011). (in Vietnamese) 16. Bozanic, D., Milic, A., Tesic, D., Sałabun, W., Pamucar, D.: D numbers – fucom – fuzzy rafsi model for selecting the group of construction machines for enabling mobility. Facta Univ. – Mech. Eng. 19(3), 447–471 (2021). https://doi.org/10.22190/FUME210318047B 17. Muhammad, L.J., Badi, I., Haruna, A.A., Mohammed, I.A.: Selecting the best municipal solid waste management techniques in nigeria using multi criteria decision making techniques. Rep. Mech. Eng. 2(1), 180–189 (2021). https://doi.org/10.31181/rme2001021801b 18. Le, H.-A., Hoang, X.-T., Trieu, Q.-H., Pham, D.-L., Le, X.-H.: Determining the best dressing parameters for external cylindrical grinding using MABAC method. Appl. Sci. 12(16), 8287 (2022). https://doi.org/10.3390/app12168287 19. Pamucar, D., Behzad, M., Bozanic, D., Behzad, M.: Decision making to support sustainable energy policies corresponding to agriculture sector: case study in Iran’s Caspian Sea coastline. J. Clean. Prod. 292, 125302 (2021). https://doi.org/10.1016/j.jclepro.2020.125302

Combining DOE and EDAS Methods

227

20. Trung, D.D., Thinh, H.X.: A multi-criteria decision-making in turning process using the MAIRCA, EAMR, MARCOS and TOPSIS methods: a comparative study. Adv. Prod. Eng. Manage. 16(4), 443–456 (2021). https://doi.org/10.14743/apem2021.4.412 21. Keshavarz-Ghorabaee, M., Amiri, M., Zavadskas, E.K., Turskis, Z., Antucheviciene, J.: Determination of objective weights using a new method based on the removal effects of criteria (MEREC). Symmetry 13(4), 525 (2021). https://doi.org/10.3390/sym13040525 22. Trung, D.D., Nguyen, N.-T., Duc, D.V.: Study on multi-objective optimization of the turning process of EN 10503 steel by combination of Taguchi method and Moora technique. EUREKA: Phys. Eng. 2, 52–65 (2021). https://doi.org/10.21303/2461-4262.2020.001414 23. Trung, D.D.: A combination method for multi-criteria decision making problem in turning process. Manuf. Rev. 8(26), 1–17 (2021). https://doi.org/10.1051/mfreview/2021024 24. Trung, D.D., Thien, N.V., Nguyen, N.T.: Application of TOPSIS method in multi-objective optimization of the grinding process using segmented grinding wheel. Tribol. Ind. 43(1), 12–22 (2021). https://doi.org/10.24874/ti.998.11.20.12

Testing the Durability of the Color of Façade Materials Piotr Kosi´nski(B)

and Agata Jodko

University of Warmia and Mazury in Olsztyn, ul. Heweliusza 10, 10-724 Olsztyn, Poland [email protected]

Abstract. The paper presents a problem of discoloration of facade materials. 10 different materials, including acrylic resins and paints, cement mortars, and synthetic membranes were tested in accelerated weather conditions to obtain their colors’ durability. The climatic conditions were generated by the accelerated weathering tester, QUV. The research on color durability was made with the Minolta CM-700d spectrophotometer. The research was divided into two stages. In the first stage, three factors, including moisture spray, condensation, and UVA acted on the materials, while in the second stage, only UVA radiation acted on the materials. Materials differ in discoloration results. For acrylic and cement one combination of climate conditions (with an indication of moisture) seem to be the driving discoloration factor. In the case of membranes, UVA is no less color damaging. Only in a few materials, including all cement ones, the discoloration was so big, that inexperienced observers can see them. Keywords: Discoloration assessment · Building plasters and paintings · Building aesthetics

1 Introduction 1.1 General Information Nowadays, great importance is attached to the aesthetics of buildings and the general care for their finish. Even an inexperienced observer notices defects caused during long-term use of the object. The durability of the color of the materials is no less important than their structural durability expressed, for example, by resistance to mechanical damage. Mostly, the defects and damages are accelerated by climate factors. Solar radiation is one of the main climate exposure factors by which many building materials may be affected and even damaged. Solar radiation may initiate photodegradation which is the leading mechanism of discoloration of facade materials [1]. Water in various states is another important, climatic factor that influences the color durability of materials. Water might even start the leading process of degradation of façade elements when washing out the small elements from materials [2]. Materials made of natural stone or ceramics prove the highest resistance to discoloration. However, modern finishes, especially façades, are largely made of synthetic © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 228–240, 2024. https://doi.org/10.1007/978-3-031-51494-4_20

Testing the Durability of the Color of Façade Materials

229

materials. These materials are less resistant to weather conditions and they are more susceptible to discoloration or mechanical damage [3, 4]. As a result, buildings lose their aesthetic values. The paper describes tests performed for 10 different synthetic materials used in building facades. Acrylic resins and protective paints, mortars, and waterproofing materials were tested in an accelerated weather tester to assess their color durability. The colors of the tested materials varied, depending on the manufacturer. The colors’ durability was expressed as spectral changes presented on the CIE L*a*b scale recorded by a spectrophotometer. 1.2 Previous Works Dedicated to Color Durability of Building Materials Research on the durability of the color of external building materials is carried out in many ways. Some of these works concern discolorations that may appear with the aging of the material and are intensified by climatic factors. Mainly, façade materials made of stone [5–9], ceramics [10–13], wood [14–17], or metal [18] are subject to testing. In the case of wood elements, both rough or impregnated materials are tested. Some research is carried out in situ and then includes long-term observations. Some are laboratory works supported by accelerated weather testers. Although tests are carried out on different materials by different research centers, the conclusions are similar, there are no materials that are not subject to discoloration. So far, no common methods of accelerated aging of materials are uniform. Mainly researchers use their own experience. In the case of wood and wood-based materials, Scandinavian experience is included in Nordtest Method NT Build 495 [19]. Artificial aging tests for flexible waterproofing products for roofing are included in the EN 1297 standard. However, apart from the suggested time of heating and moistening the samples, EN 1297 does not provide procedures for assessing changes under the influence of accelerated weathering, suggesting the observation of changes in appearance, weight, and crack formation. There are also works on coatings ensuring durability, including protection against discoloration [7, 15, 17, 20]. They conclude that the coatings require renewal, but darker colors of semi-transparent coating systems lead to better protection of the substrate against harmful wavelengths of light. Authors find also that Artificial cyclic exposures tend to be more severe than natural exposures, but on the other hand, they overlook the damage of bio-deterioration. Color-fastness tests of other materials used outside buildings are rare and are usually presented in the form of product advertisements.

2 Materials and Methods 2.1 Materials Various external coating materials (Table 1) dedicated to exterior finishing were the object of the research. These were coatings made of elastic acrylic resins, acrylic protective paints, sealing materials made of cement and synthetic polymers, and waterproofing membranes in PVC-P or flexible polyofin. The colors of the materials were various. The materials were applied to thin fiberboards measuring 25 × 7 cm. Two samples were

230

P. Kosi´nski and A. Jodko

made for each material. Before starting the tests, i.e. inserting the tested samples into the accelerated weather tester, each sample surface was thoroughly cleaned with distilled water. Table 1. List of tested external coating materials.

¸No.

1 2 3 4 5 6 7 8 9 10

Color (CIE L*a*b) 78.66*-1.38*1.62

Material description Acrylic resin

33.16*-1.48*-2.16

Acrylic resin

92.00*-0.75*3.61

Acrylic protective paint

78.81*-1.34*1.75

Acrylic protective paint

38.16*-4.55*9.94

Acrylic protective paint

40.82*1.41*9.62

Flexible cement mortar

39.20*3.60*10.42

Sealing based on cement and synthetic polymers

42.47*0.84*11.71

Sealing based on cement and synthetic polymers

95.98*-0.90*3.88

waterproofing membrane in flexible polyolefin

80.26*-0.91*1.91

waterproofing membrane in PVC-P

2.2 Methods Color Spectrophotometric Measurements The research consisted in measuring the color spectrum of samples in their original state and after aging in an accelerated weather tester QUV. Spectral measurements of the samples were made using a portable Minolta CM-700d spectrophotometer. The spectrophotometer uses built-in light of known spectral composition to test color. In case of the device used, it is a pulsed xenon lamp with a filter that cuts off UV radiation. The measuring system includes an integrating sphere with a diameter of 40 mm, a spectrum detector in the form of a matrix of silicon photodiodes consisting of 36 double elements, and a diffraction grating used to split the light. Measurement results are presented in spectral form. However, they can be presented in better-known color spaces, e.g. CIE*Lab, CIE*Lch, CIE*XYZ, and Hunter Lab. The work uses the CIE*Lab scale, which is one of the most accurate scales when analyzing color differences [21]. Before each use of the spectrophotometer, it had to be calibrated using the white standard. Then the illuminant and the observer had to be set up. An illuminant D65 was selected - daylight with UV content with a color temperature of 6504 K and an observer at 10◦ , which corresponds to the conditions of the measurements. All obtained color change results are averages of the four extreme corners of the sample. This made it possible to eliminate inaccuracies in the examination, consisting of the unintentional shift of the tested point. The measurements were performed without the specular component (reflection) SCE (Specular Component Excluded).

Testing the Durability of the Color of Façade Materials

231

With the values of the CIE Lab scale, it is possible to compare both individual coordinates and the difference Eab (given with the formula 1) expressing the Euclidean distance of individual values of two samples in the three-dimensional color space.  ΔEab = (ΔL)2 + (Δa)2 + (Δb)2 (1) where: Eab - color difference L - luminance value difference, a - position difference between green and red color b - position difference between blue and yellow. Based on the statistical data, it can be assumed that an experienced observer is able to see the difference in color of two objects with Eab ≥ 1.0, while an inexperienced observer will notice a difference of at least Eab = 3.5. Accelerated Aging of the Materials In the accelerated weather tester QUV (Fig. 1) it was possible to set cycles of UV radiation, condensation, and water spray. Fluorescent lamps simulating the effect of UVA radiation, which is the closest to the radiation of sunlight, were used for the research. The research was divided into two stages. In the first stage, the load on the samples was as follows: rain (20 min), water vapor condensation at 45 ◦ C (13 h 40 min), and UVA irradiation at 0.68 W/m2 (10 h). This cycle was repeated automatically with only breaks for sample discoloration control measurements. The second stage consisted of continuous irradiation of the samples with UVA radiation at 0.68 W/m2 . In the second stage, the samples tested in the first stage were left in the QUV, and samples of the same materials that were previously stored in the darkroom were added. The purpose of dividing the research program into two stages was to enable a comparative analysis of the impact of individual mechanisms on the discoloration of materials. Before the launch of QUV, spectrophotometric measurements of samples of the tested materials were performed. These results were set as a pattern for further analysis. During the first stage of the research, measurements were made with a spectrophotometer after 10 full cycles, 16 and 23. Each cycle lasted a total of 24 h. During the second stage of the research, spectrophotometric measurements were made after 7 days, 15, 21 and 28. The authors intended to perform measurements at intervals of every 7- full cycles. Unfortunately, for reasons beyond the authors’ control, some measurements were moved in time. Hence, apart from the first reading, which occurred only after 10 cycles, the interval was maintained.

232

P. Kosi´nski and A. Jodko

Fig. 1. One of the tested materials (a), a complete of materials in the QUV apparatus (b).

3 Results and Discussion Tables 2, 3, 4, 5 and 6 present a summary of the results for two measurement series (rain + condensation + UVA and only UVA) for two samples of each tested material. For each of the presented samples, the values in the CIE*Lab scale are given for the material in its original state, which was considered as a pattern. Next, the values of the CIE*Lab components of the samples loaded with rain + condensation + UVA are read after 10, 16, and 23 days. Then, for all samples, CIE*Lab component values after additional 7, 15, 21, and 28 days of UVA-only loading. Such a combination facilitates the analysis of the influence of UVA-coupled moisture factors and UVA itself on the discoloration of the tested coating materials. For all measurements of weather-loaded samples, Eab values are also presented, later referred to as discoloration. It should be noted that spectrophotometric measurements of samples of the same material in the original state showed slight differences, below 1%, between the L*a*b components, which means that the materials are not perfectly uniform in color. This confirms the correct selection of the method of presenting the average of the readings taken at four fixed points of the samples. For both acrylic resins shown in Table 2, there is a trend that the first 10 days of moisture and UVA exposure resulted in the greatest discoloration. It was caused mainly by a decrease in the L component, i.e. darkening of the surface, and in the 1st material, an additional increase in the b component. Eab ≥ 1.80. Further moisture and UVA loading did not bring any major discoloration changes in the 1st material, but the 2nd material shows an increase in the L component compared to previous readings, and thus a decrease in Eab. Loading the materials with UVA alone caused a slight discoloration of material 1 (decrease in component b), while in the 2nd material, the effect was comparable to the effect of moisture and UV. It can be seen that for the 1st material previously exposed to moisture and UVA, further UVA irradiation resulted in a decrease in discoloration, mainly by an increase in the L component and a decrease in the b component. For the 3rd material (Table 3), the moisture and UVA loading of the sample resulted in changes in Eab ≥ 1.20, mainly by reducing the L and b components. Further UV loading of this sample resulted in a decrease in discoloration, mainly by an increase in the L component. UVA loading alone resulted in discoloration Eab ≥ 1.07 caused mainly by an increase in the L component and a decrease in the b component. For the 4th material (Table 3),

Testing the Durability of the Color of Façade Materials

233

Table 2. The results of the color spectrum measurements, presented in the CIE*Lab scale for acrylic resins. pattern

10 days

16 days

23 days

23 + 7 days

23 + 15 days

23 + 21 days

23 + 28 days

L

78.66

77.41

77.46

77.5

77.64

77.64

77.69

77.78

a

−1.38

−1.27

−1.25

−1.21

−1.24

−1.27

−1.33

−1.29

b

1.62

3.05

3.09

2.99

2.39

2.02

ΔEab

−

1.91

1.90

1.80

1.29

1.10

1.08

L

78.29

78.48

78.31

78.32

78.3

a

−1.36

−1.34

−1.34

−1.36

−1.34

b

1.66

No. 1st moist & UV

only UV

ΔEab

2.1

1.95 0.95

1.30

1.22

1.34

1.24

0.40

0.45

0.32

0.42

2nd moist &UV

L

33.16

31.77

31.86

32.13

31.99

31.8

31.86

32.03

a

−1.48

−1.64

−1.63

−1.62

−1.63

−1.65

−1.64

−1.63

b

−2.16

−2.36

−2.34

−2.31

−2.27

−2.26

−2.27

−2.22

1.42

1.32

ΔEab only UV

1.19

1.37

1.32

1.14

L

33.25

32.88

32.34

32.33

32.08

a

−1.53

−1.52

−1.62

−1.56

−1.62

b

−2.17

−2.14

−2.11

−2.1

−2.04

0.37

0.91

ΔEab

1.05

0.92

1.18

the greatest discoloration occurred when exposed to moisture and UVA (decrease in L, increase in a and b), and it is increasing with time, Eab ≥ 3.27. Further UVA loading resulted in a steady decrease in discoloration (increase in L, decrease in a and b) over time.

234

P. Kosi´nski and A. Jodko

Table 3. The results of the color spectrum measurements, presented in the CIE*Lab scale for acrylic protective paints. pattern

10 days

16 days

23 days

23 + 7 days

23 + 15 days

23 + 21 days

23 + 28 days

L

92

90.88

90.7

90.86

91.07

91.11

90.76

91.19

a

−0.75

−0.46

−0.43

−0.39

−0.55

−0.55

−0.61

−0.68

b

3.61

3.91

3.85

3.53

3.28

3.22

3.63

3.33

1.20

1.36

1.20

No. 3rd moist & UV

ΔEab only UV

1.01

0.99

1.25

0.86

L

91.93

92.18

92.17

92.12

92.15

a

−0.75

−0.74

−0.73

−0.79

−0.77

b

3.8

ΔEab

2.34

2.23

2.74

2.3

1.49

1.59

1.07

1.51

4th moist &UV

L

78.81

76.78

76.76

76.76

77.65

77.92

77.99

78.2

a

−1.34

−1.01

−0.94

−0.81

−0.97

−1.06

−1.12

−1.11

b

1.75

3.22

2.62

2.61

2.37

ΔEab only UV

4.3

4.49

4.45

3.27

3.45

3.43

1.2

0.89

L

77.98

78.16

78.1

78.06

78.12

a

−1.29

−1.28

−1.28

−1.31

−1.28

b

1.82

ΔEab

1.91

1.27

1.55

1.48

1.62

1.55

0.32

0.37

0.22

0.30

5th moist &UV

L

38.16

37.77

37.76

37.53

37.65

37.57

37.6

37.53

a

−4.55

−4.39

−4.37

−4.36

−4.33

−4.25

−4.23

−4.09

b

9.94

ΔEab only UV

9.74

9.8

10.06

9.86

9.5

9.5

9.38

0.46

0.45

0.66

0.56

0.78

0.77

0.95

L

38.33

38.08

37.81

37.86

37.7

a

−4.52

−4.52

−4.52

−4.48

−4.4

b

10

10.16

10.24

10.16

10.2

0.30

0.58

0.50

ΔEab

0.68

The UVA load alone caused a virtually permanent discoloration in the material, mainly by a slight increase in L and a decrease in b. For the 5th material, the discoloration due to moisture and UV load coincides with the UVA load alone, being mainly caused by a decrease in L and an increase in b. Continuation of the UVA load of the first sample

Testing the Durability of the Color of Façade Materials

235

of the 5th material (Table 3) causes a constant increase in discoloration, but invisible to the naked eye. Table 4. The results of the color spectrum measurements, presented in the CIE*Lab scale for flexible cement mortar. pattern

10 days

16 days

23 days

23 + 7 days

23 + 15 days

23 + 21 days

23 + 28 days

L

40.82

52.44

55.23

56.04

57.66

64.03

64.05

70.48

a

1.41

0.58

0.57

0.56

0.52

0.17

0.10

−0.06

b

9.62

5.24

5.52

5.4

4.69

3.26

3.34

2.57

15

15.81

No. 6th moist & UV

ΔEab only UV

17.57

24.1

24.1

30.52

L

41.01

41.48

45.2

40.36

50.53

a

1.53

1.41

0.98

1.31

0.63

b

9.91

9.99

8.6

10.08

7.35

0.48

4.42

0.71

9.89

ΔEab

12.45

Moisture and UVA loading of mortar cement resulted in large changes in ΔEab > 12.45 with a steadily increasing trend (Table 4). This is caused by a significant increase in the L component, a reduction in the a and b components. Further UVA loading of this sample resulted in a continuation of the trend and an increase in ΔEab to 30.52 after 28 days. Table 5 presents the CIE L*a*b results and ΔEab difference with patterns of measurements for two sealing based on cement and synthetic polymers. For both materials, discoloration caused by moisture and UVA or UVA alone is increasing over time. For the 7th material, there is mainly an increase in the a and b components, the L component increases primarily under the influence of UVA. Changes caused by different climatic load modes are comparable over their duration. Color changes are visible to the naked eye (ΔEab ≥ 4.65). The discoloration is greater for the 8th material. Here, the changes caused by moisture and UVA are dominant, mainly an increase in the L component and a gradual reduction of a and b. During the load with only UVA, the increase of L is mainly observed. Table 5 presents the CIE L*a*b results and ΔEab difference with patterns of measurements for two waterproofing membranes. The effect of accelerated weathering on the 9th material is visible as rising ΔEab in time. The effect of UVA is comparable to combined moisture and UVA. The discoloration is caused primarily by increasing the b component. The discoloration of the 10th material is practically negligible regardless of the research mode, ΔEab < 0.2. Figure 2 presents the results of the investigation on the discoloration of façade materials subjected to climate load generated by an accelerated weather tester. The individual graphs show a comparison of two generated conditions, spray and condensation with a combination of UVA, and only UVA. Each material behaved differently. Figure 2a and

236

P. Kosi´nski and A. Jodko

Table 5. The results of the color spectrum measurements, presented in the CIE*Lab scale for sealing based on cement and synthetic polymers. pattern 10 days 16 days 23 days 23 + 23 + 23 + 23 + 7 days 15 days 21 days 28 days

No. 7th moist L & UV a b

39.2

39.24

3.60

4.48

4.54

4.66

5.04

5.18

5.18

5.08

10.42

12.72

12.81

12.85

13.58

14.67

14.71

14.53

2.47

2.58

2.66

ΔEab only UV

39.36

39.15

38.57

39.44

39.32

41.68

3.54

4.55

4.57

5.02

L

39.24

39.08

38.6

38.3

42.3

a

3.55

3.93

4.29

4.3

4.24

b

10.21

10.98

12.67

12.69

13.65

0.87

2.65

2.75

4.65

ΔEab 8th moist &UV

L

42.47

53.83

55.1

56.07

58.85

62.19

62.06

64.99

a

0.84

0.58

0.55

0.53

0.4

0.21

0.16

0.12

b

11.71

10.19

10.08

9.64

8.79

8.12

8.28

7.32

11.46

12.74

13.75

ΔEab only UV

16.64

20.05

19.9

22.95

L

42.58

43.06

46.76

46.01

50.32

a

0.90

1.01

0.73

0.87

0.7

b

11.75

12.89

11.69

12.45

11.09

1.24

4.18

3.5

7.77

ΔEab

2b present the discoloration obtained for acrylic resins, lighter and darker, respectively. It is visible that moisture is a leading discoloration force for lighter resin, while for darker UVA plays a significant role. What is more, the discoloration of lighter resin loaded with combined climate conditions decreases when treated further only with UVA. The discoloration obtained within almost 2 months of accelerated aging of materials can be seen only by experienced observers. Figure 2c–e present the discoloration obtained for acrylic protective paints, from the lightest to the darkest, respectively. In the case of the lightest, UVA proves to be the leading damaging force. In the case of medium light paint, moisture plays a significant role in discoloration, while only UVA load decreases the changes when further operates. Moisture damages can be seen by inexperienced observers as well. In the case of the darkest material discoloration is almost not visible, and both loads’ combinations resulted similarly, with an indication of UVA as the leading factor. Figure 2f presents the discoloration of cement mortar. Spray and condensation are the driving forces of the damage. UVA plays a smaller role, but even if the material were treated with UVA only, the discoloration is visible to inexperienced observers.

Testing the Durability of the Color of Façade Materials

237

Table 6. The results of the color spectrum measurements, presented in the CIE*Lab scale for sealing based on waterproofing membrane in flexible polyolefin, or PVC-P. pattern 10 days 16 days 23 days 23 + 23 + 23 + 23 + 7 days 15 days 21 days 28 days

No. 9th moist L & UV a b

95.98

95.49

95.48

95.44

95.5

95.43

95.31

95.42

−0.9

−0.76

−0.75

−0.73

−0.88

−0.9

−1.02

−0.98

3.88

4.86

4.73

4.9

5.02

5.14

5.89

5.42

1.11

1

1.17

1.24

1.37

2.13

1.64

95.08

94.96

94.82

94.84

−0.85

−0.89

−1.08

−0.89

5.37

5.75

6.67

6.07

ΔEab only UV

L a

94.69

b

0.27

ΔEab 5.58 10th moist &UV

L

80.25

80.33

80.34

80.41

80.39

80.32

80.29

80.44

a

−0.91

−0.9

−0.9

−0.89

−0.89

−0.88

−0.89

−0.88

b

1.99

1.94

1.93

1.97

1.98

1.95

2.01

1.95

0.09

0.11

0.16

ΔEab only UV

0.14

0.09

0.05

0.19

L

80.26

80.21

80.18

80.16

80.24

a

−0.92

−0.92

−0.91

−0.93

−0.89

b

1.96

2.08

2.07

2.18

2.04

0.13

0.13

0.25

0.09

ΔEab

Figure 2g and h present discoloration of sealing based on cement and synthetic polymers, both similar in color darkness. Similar to cement mortar, spray and condensation are the driving forces of discoloration. Similarly, UVA plays a smaller role, but even if the material were treated with UVA only, the discoloration is visible to inexperienced observers. Figure 2i and j present discoloration of waterproof membranes, lighter and darker, respectively. In the case of a lighter one, the load effect is similar for moisture combined with UVA and only UVA. The discoloration can be visible only for experienced observers. The discoloration of the darker membrane is almost not visible, with UVA as a driving force.

238

P. Kosi´nski and A. Jodko

Fig. 2. List of discoloration (Eab) for individual materials treated with spray + condensation + UV or only UV, a, b) acrylic resins, c, d, e) acrylic protective paints, f) cement mortar, g, h) sealing based on cement and synthetic polymers, i, j) waterproofing membranes.

4 Conclusion The research aimed to determine the durability of the color of façade materials that were exposed to external factors: UVA radiation, rain, and water vapor condensation. The research was divided into two stages. In the first stage, all three factors acted on the materials, while in the second stage, only UVA radiation acted on the materials. The materials showed different resistance to climatic factors. The greatest discoloration can be observed for cement-based materials. For them, the leading discoloration factor is a combination of rain, condensation, and UVA. A high level of moisture causes the particles to be washed out from between the particles of the material, causing visible color changes. The UVA itself is also damaging to them, but not to such a great extent. In

Testing the Durability of the Color of Façade Materials

239

acrylic materials, the effect of moisture in combination with UVA also resulted in greater discoloration (except for light acrylic paint). However, it should be borne in mind that color changes in acrylic materials during the research period were not as great as in the case of cement ones. Moreover, for several of them, a subsequent single UVA treatment reduced the discoloration of samples loaded with the full climatic combination. In the case of membranes, greater discoloration occurred for flexible polyolefins, rather than for PVC-P. However, it cannot be clearly stated whether this is the effect of material resistance or the original color of the samples. The vast majority of discoloration of the tested materials resulted from a change in the L (brightness) component. The presented results are cognitive and can be used for further work on the durability of the color of external materials used in buildings, subjected to moisture and UVA.

References 1. Jelle, B.P.: Accelerated climate ageing of building materials, components and structures in the laboratory. J. Mater. Sci. 47, 6475–6496 (2012) 2. Kosi´nski, P., Parzych, P., Dom˙zalski, F., The silicone plaster water absorptivity under climate load. In: AIP Conference Proceedings, vol. 2801, no. 1, p. 030017 (2023). https://doi.org/10. 1063/5.0146727 3. Viegas, C.A., Borsoi, G., et al.: Diversity and distribution of microbial communities on the surface of external thermal insulation composite systems (ETICS) facades in residential buildings. Int. Biodeterior. Biodegrad. 184, 105658 (2023) 4. Parracha, J.L., Borsoi, G., et al.: Performance parameters of ETICS: correlating water resistance, bio-susceptibility and surface properties. Constr. Build. Mater. 272, 121956 (2021) 5. Lisci, C., Sitzia, F., et al.: Building stones durability by UVA radiation, moisture and spray accelerated weathering. J. Build. Pathol. Rehabil. 7, 60 (2022) 6. Sitzia, F., Lisci, C., Mirao, J.: Accelerate ageing on building stone materials by simulating daily, seasonal thermo-hygrometric conditions and solar radiation of CSA Mediterranean climate. Constr. Build. Mater. 266, 121009 (2021) 7. Lopez-Arce, P., Tagnit-Hammou, M., et al.: Durability of stone-repair mortars used in historic buildings from Paris. Mater. Struct. 49(12), 5097–5115 (2016) 8. De Kock, T., Dewanckele, J., et al.: Replacement stones for lede stone in Belgian historical monuments. Geol. Soc. Spec. Pub. 391(1), 31–46 (2014) 9. Benavente, D., Martinez-Verdu, F., et al.: Influence of surface roughness on color changes in building stones. Color. Res. Appl. 28(5), 343–351 (2003) 10. Cultrone, G., De la Torre, M.J., et al.: Behavior of brick samples in aggressive environments. Water Air Soil Pollut. 119, 191–207 (2000) 11. Pérez-Monserrat, E.M., Maritan, L., et al.: Production technologies of ancient bricks from Padua, Italy: changing colors and resistance over time. Minerals 11, 744 (2021) 12. Pérez-Monserrat, E.M., Agua, F., et al.: Effect of manufacturing methods on the decay of ceramic materials: a case study of bricks in modern architecture of Madrid (Spain). Appl. Clay Sci. 135, 136–149 (2017) 13. Eramo, G.: Ceramic technology: how to recognize clay processing. Archaeol. Anthropol. Sci. 12(8), 1–24 (2020). https://doi.org/10.1007/s12520-020-01132-z 14. Grüll, G., Tscherne, F., et al.: Comparison of wood coating durability in natural weathering and artificial weathering using fluorescent UV-lamps and water. Eur. J. Wood Wood Prod. 72, 367–376 (2014) 15. Zhang, L., Yang, X., et al.: Properties and durability of wood impregnated with high melting point polyethylene wax for outdoor use. J. Wood Chem. Technol. 42(5), 342–351 (2022)

240

P. Kosi´nski and A. Jodko

16. Srinvas, K., Pandey, K.K.: Effect of heat treatment on color changes, dimensional stability, and mechanical properties of wood. J. Wood Chem. Technol. 32(4), 304–316 (2012) 17. Kubovsky, I., Kacik, F.: Changes of the wood surface colour induced By CO2 laser and its durability after the xenon lamp exposure. Wood Res. 58(4), 581–590 (2013) 18. Ihara, T., Jelle, B.P., et al.: Accelerated aging of treated aluminum for use as a cool colored material for facades. Energy Build. 112, 184–197 (2016) 19. Nordtest Method NT Build 495 (2000) Building materials and components in the vertical position: exposure to accelerated climatic strains 20. Shirakawa, M.A., Loh, K., et al.: Biodeterioration of painted mortar surfaces in tropical urban and coastal situations: comparison of four paint formulations. Int. Biodeterior. Biodegradation 65(5), 669–674 (2011) 21. Lopez, F., Valiente, J.M., et al.: Fast surface grading using color statistics in the CIE lab space. In: Marques, J.S., de la Blanca, N.P., Pina, P. (eds.) IbPRIA 2005. LNCS, vol. 3523, pp. 666–673. Springer, Heidelberg (2005). https://doi.org/10.1007/11492542_81

Dimensional Measuring System with Temperature Compensation Nemanja Zuji´c1 and Djordje Dihovicni2(B) 1 Gazela d.o.o, Krško, Slovenia 2 Department of Computer-Mechanical Engineering, Academy of Technical Applied Studies

Belgrade, Belgrade, Serbia [email protected]

Abstract. This paper presents the development and creation of a system for measuring length with temperature compensation. The measuring system serves for the final dimensional control of the quality of the product. It is a necessary condition to ensure the conformity of the product with the dimensional values given in the technical drawing. The solution for the case of the product condition with temperature expansion is a measuring system, which has integrated temperature compensation. This means that every time, in addition to the measurement, it also measures the temperature of the product and takes into account the appropriate correction factor for the calculation of the actual value. This allows user to obtain standardized measurement results independent of the temperature of the product at the time of measurement. Keywords: Temperature compensation · control systems · industrial computer · measuring system

1 Introduction This research describes the development and fabrication of a temperature compensated length measurement system, [1, 2]. The measuring system serves for the final dimensional control of the quality of the product, which comes from machining, [3–6]. It is a necessary condition to ensure the conformity of the product with the dimensional values given in the technical drawing, [7–9]. Value deviations are allowed only within the measurement tolerance interval of each measured characteristic, [10–13]. The ability to ensure process stability is a condition for the success and competitiveness of companies, especially those that mechanically produces large series of products, [14–17]. The measuring system consists of an industrial computer that processes the data and displays it on the LCD screen, a temperature sensor, which reads the product temperatures and sends it to the computer, an input/output control unit to read the status, a foot switch, a light indication to indicate the measuring pins, an inductive probe which are housed in dedicated micron precision hole, diameter measuring heads and the Millimar X1715 analog/digital interface, manufactured by Mahr. Combining theoretical and practical knowledge in the field of automation, mechanical constructions, thermo technical © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 241–258, 2024. https://doi.org/10.1007/978-3-031-51494-4_21

242

N. Zuji´c and D. Dihovicni

and ventilation systems, with application standards and regulations are systematically improved the management of this responsible technical system, [18–23]. With a measuring system that has software built-in temperature compensation, it is enabled to the user to measure diameters of the hole, with an accuracy of 1 µm or 1/1000 of a millimeter, independent of the processed product temperature. The final control of the product in serial machining is a very important process to ensure the production quality of the product itself. Control preparations are generally used where the permitted tolerance deviations are so small that they cannot be controlled with ordinary measuring instruments, [24–26]. With the final control, it is checked not only the quality of the processing, but also deviations from the nominal dimensions within the specified interval of the permitted tolerance field and thus the reduction of deviations, which are a consequence of the wear of the cutting tools of CNC processing machines, i.e. external influences. Measuring systems are designed based on the drawn three-dimensional plan of the model and prescribed dimensions, as well as product tolerances. The user creates a list of characteristics that need to be controlled and based on all this information, a concept solution and implementation of a control or measuring device. The attributive measures of devices and measuring devices for measuring given dimensions are known. With attributive control of preparations or calibers, product reference points are controlled according to the corresponding procedure (GO/No GO). Actual dimensions and geometry are measured with a measuring device based on sensors, calculated deviations from the nominally stated dimensions and measurement protocols. A common problem with measuring systems is the final control in CNC machining of heated hot products. This is especially important in the case of aluminum products, which have a large temperature expansion, which means that they can form large value deviations. One solution would be to have machined parts waiting in the gauge. Taking into account that it is usually a large-scale production, the final inspection should be done when the product leaves the processing machine. The solution for the case of the product condition with temperature expansion is a measuring system, which has integrated temperature compensation. This means that every time, in addition to the measurement, it also measures the temperature of the product and takes into account the appropriate correction factor for the calculation of the actual value. This allows mechanics to obtain standardized measurement results independent of the temperature of the product at the time of measurement. The measuring system must perform the following basic tasks: • • • • • • •

reading the prescribed hole diameters according to the customer’s requirements; reading the temperature and calculating the temperature compensation for the product; standards; display of measurements on the LCD screen; recording of measurements in the archive of the history of measurement results; managing measurements with a foot switch; light indication for the selection of measuring pins.

During the development and production of the measuring system, the focus is on its production costs. Therefore, components that are relatively cheap and suitable for them and the production of the entire system were searched for and selected. The goal of the

Dimensional Measuring System with Temperature Compensation

243

task is to create a system that will be reliable and will perform all the functions it must perform according to the given requirements and influences.

2 Description of the Measurement System and Operation The task of creating a measuring system with temperature compensation was requested by the client LTH Alucast Company Gazela d.o.o. from Krško, which deals with the production and realization of measuring devices of the system. From Gazele d.o.o. it was found that a system containing temperature compensation can be measured, avoiding the problem of incorrect measurements due to high temperature in production lines and products. Through the realization and development of this project, a whole process will be introduced, from the very beginning, from the problem to the idea and the complete solution. During operation and testing of the entire measuring system and possible errors during temperature compensation, the system provides remote access to the computer and in these way errors can be corrected quickly. In addition to solving problems, remote access also allows control, whether the worker or the user is using the measuring system correctly. The main features of the system are: • • • • • • • • • •

Mechanical movements of the measuring heads to the inductive probe P2004M; Measurement of hole dimensions with the Millimar X1715 measuring interface; Non-contact infrared temperature measurement; Measuring system with temperature compensation; Data processing, measurement process control and communication with peripheral devices are handled by the ICO300 industrial computer; Foot switch for recording measurements and controlling the measuring system; LED indication for easier selection of the measuring head during the measurement step, Graphic display of the measuring step and the value of the measured dimensions of the hole on the LCD screen, 24 V power supply via switching power supply, Measuring cabin with light, work counter and stand for measuring heads facilitates measurement, product manipulation and protection of the measuring system against external influences.

The mechanical measuring head, which has a built-in electronic inductive probe, has the role of reading changes in the dimension of the measured hole. The expected changes are on the micron level, so the mechanics need to use an inductive probe that has a resolution of 0.01 µm and an accuracy of 0.1 µm. A voltage change appears at the output of the inductive probe. This voltage change is processed by the Millimar x1715 measuring unit, which sends digital data via RS232 to the computer. The computer processes the data and displays the measured value on the screen. A dedicated industrial computer was used for the core of the measuring system ICO300, manufactured by Akiomtek, which is mounted on a DIN rail. The advantage of such a computer is that it does not have a fan for forced cooling of the processor,

244

N. Zuji´c and D. Dihovicni

because it has a built-in small Intel Atom processor, which works up to 1.46 GHz, which is enough for the operation of Windows applications. Such a computer is very convenient in systems where the concentration of moisture or dust is very high, which is very common in industry. The computer is made of a robust small metal case that can be mounted in various positions on a DIN rail. It is designed for ambient temperatures between −20 °C and up to 70 °C. The power supply is connected via a 24 V port, and the display is connected to a standard VGA connector. The connectors are located on the top of the computer. In addition to the basic connections, there are 4 serial ports built into the front side of the communication port, where it can be chosen between three communication protocols: RS232, RS422 and RS485. Two LAN ports for PROFINET communication or connection to the Internet for remote access are also built in. The percentage of moisture is in all technological processes as well as in the treatment of living space, and that must be taken into account, because, it leads to very large change in system state. Built-in Industrial Computer Features: • • • • • • • • • • • • • • • • • • • • • • • • •

Passive cooling; High ambient temperature −20 ◦ C to 70 ◦ C; Ambient humidity between 10% and 95%; Two magnetically isolated Ethernet ports up to 1 Gb/s; 4 COM serial ports with support for RS232/422/485 protocol; Support for GSM or Wi-Fi communication; Supports 2.5 SATA drive, CompactFlash™ or mSATA; Supply voltage between 12–24 VDC; Mounting on a DIN rail; Support for embedded systems such as Windows or Linux systems. System Specifications: Built-in Intel™ ATOM™ processor marked E3815 (1.46 GHz); Working RAM DDR3 support up to 4 GB; 15 Pin D – Sub VGA connector; Two RJ-45 connectors for Ethernet; Data storage on a 2.5 SATA disk, CompactFlash™ or mSATA; SIM card support; Wi-Fi network support; Two USB 2.0 connectors; Two holes for antennas; Four COM DB9 connectors for serial communication; Power supply connector for a voltage between 12–24 V; Watchdog timer (VDT); The possibility of operating the system at optimal settings; Two system LEDs: ACT for disk activity and PWR for indication of power supply voltage; • Weight: 1 kg; • Dimensions: 48 × 110 × 155 mm.

Dimensional Measuring System with Temperature Compensation

245

The measuring head BMD in two points, manufactured by DIATEST, is a precise measuring device tool for measuring the diameter of the opening. Two-point measuring heads are very easy to use and suitable for static and dynamic measurements. They can be used as individual gauges to measure hole diameters with an additional gauge or as an angle component of an automated batch measuring machine. They have high accuracy and robust construction and are suitable for difficult measurement conditions in industrial production halls. They enable accurate diameter measurements. Display of measuring head with holders is shown in Fig. 1.

Fig. 1. Display of measuring head with holders.

Characteristics of two-point measuring heads: • • • • •

easy to use; high accuracy; easy maintenance with regular cleaning; high repeatability: ≤2 µm; linearity: 1% of the measuring range. In addition to the standard two-point measuring heads, there are also heads for:

• • • • • • • •

measurement of external diameters BMD-OD; multi-level measurement of the diameter of the borehole at different heights; for measuring square holes; with depth limiter; with built-in air nozzles, instead of balls; for measuring conical openings; for measuring the diameter of the groove in wells; for measuring the diameter of gears.

246

N. Zuji´c and D. Dihovicni

To convert the mechanical movement of the balls on the two-point head into an electrical signal, it is used a precision inductive probe from the German manufacturer Mahr. The inductive probe is built into the head holder and is in direct contact with the needle which moves when the hole diameter changes. Features of the inductive probe: • • • • • • • •

required force to move the needle 0.75 N; repeatability 0.15 µm; hysteresis 0.2 µm; linear difference 3 µm; temperature coefficient 0.15 µm/°C; permitted operating temperature −10… +80 °C; work protection according to EN60529:IP64 standard; length of connecting cable: 2.5 m.

The Millimar X1715 measurement interface, manufactured by Mahr, is an intelligent digital/analog measurement interface that connects inductive probes and sends data to a computer or industrial controller. Its task is to control inductive probes and accurately read the output voltage, which is then processed and sent via RS232. In order to read the status of the foot switch and for the LED indication, a universal printed circuit board is made, whose task is to control the output and read the input. This means that according to RS232 it sends information about the change of entry in the protocol and listens to the computer. It is sort of an industrial controller as we can write a program that can run independently without computer control. The circuit itself is made in SMD technology and is installed in a plastic case that is mounted on a DIN rail. Controller is displayed in Fig. 2.

Fig. 2. Controller

Main features of the controller: • 24 V logic state reading; • 24 V outputs that provide 1 A current;

Dimensional Measuring System with Temperature Compensation

• • • • •

247

the possibility of expansion for other peripheral devices; the possibility of working independently of a computer; LED work indicator; robust RS232 communication; switching of the stabilized converter from 24 V to 5 V.

For the controller it is used Atmel’s Atmega88 integrated circuit in a TQFN package which is a low voltage CMOS 8-bit microcontroller and has an AVR RISC build. It has three input/output ports that can be used for communication and as analog inputs. The microcontroller itself works externally at a voltage of 1.8–5.5 V and a frequency of 20 MHz It has 32 × 8 operating registers controlled by Arithmetic Logic (ALE) units, which allows access to two different registers in one cycle. The V microcontroller has built-in two 8-bit timers and one 16-bit timer, which can be used for certain events in the program. For two-way communication between the computer and the controller, it is used the already known RS232 serial communication. For the logic level converter it is used the well-known MAX232 with standard connection. This connection is chosen mainly because of the stability and reliability of the communication, which is also supported by most computers, industrial controllers and peripherals. For the correct operation of the program in the AVR microcontroller, it is written a Bascom-AVR translator. It is intended for the AVR family of microcontrollers manufactured by MCS Electronics. The structure of the language itself is based on the BASIC (All Purpose Symbolic Code for Beginners) family and thus also supports the structure of the C programming language and assembly code. In automation, there are many devices that require sufficient current to operate. For this purpose it is made a controller that provides up to 1 A of current per output. When a logic unit appears at the output of the microcontroller, it is given 24 V and 1 A at the output of the controller maximum current. Since relays have a limited number of switches, it is used a P-type MOSFET for this purpose. Optical coupler SFH618A ensures that the output from the Atmega88 microcontroller is galvanic-ally isolated from the 24 V supply. It also allows not to load the output of the microcontroller. A very important role is played by the P-type MOSFET, which is connected to the plus terminal and thus provides a voltage of 24 V on the logic unit. The resistor ensures that the transistor is actually closed by R6, which is connected in a “pull-up” connection. Varistor R8 protects the MOSFET and optical spoiler from high voltage and is rated for 50 V. At 50 V it takes all the voltage and protects the transistor and the rest of the circuit before they are destroyed. Diode D5 takes care of the negative currents connected to the grounding clamp, in case of connecting a relay or other inductive loads. Relays like to cause switching interference. The advantage of the transistor output is in the speed of the switch and, of course, in the number of switches. It is used a switching power supply for power, since all the electronics in the controller run on 5 V. There would be a huge loss of power when using linear power supplies on the power supply itself due to the large difference between the input and output voltage. In our research, the input voltage of 24 V should be converted to a voltage of 5 V, which is a 19 V difference. Of course, this is not good indoors. For this purpose it is

248

N. Zuji´c and D. Dihovicni

used a LM2576 switching regulator that allows 3 A output current, which is more than enough for the described needs. The heart of the power supply is the well-known integrated circuit LM2576, which is a step-down voltage regulator, meaning it converts from a higher voltage to a lower voltage. At the input connection is possible from 7–40 V, and at the output it should be a constant 5 V. With that, it is achieved that the entire controller is made for DC voltage up to 40 V range. Since the controller needs to be universal, an additional input/output connector has been added for this purpose, to which various devices can be connected. Additional connector features: • • • • • • • •

programming via SPI; analog input; analog output; I2C digital communication; SPI digital communication; digital inputs; digital outputs; 5 V power supply.

The conditions in the machine production environment are usually very demanding and that’s why it is made a special cabin for that purpose in which it is used standard aluminum profiles. The advantage of profiles is that they can be mounted on any size or shape. For this purpose, it is used 45 × 45 mm profiles. In the cabin itself, there is a built-in light to illuminate the working environment, an electric blind, which has the role of protecting the component against oil and dust, an electrical cabinet and Plexiglas, which allow light to pass through. The measuring cabin is shown in Fig. 3.

Fig. 3. The measuring cabin

Dimensional Measuring System with Temperature Compensation

249

Characteristics of the measuring cabin: • • • • • •

protection from oil vapors and dust; separating the measuring point; lighting; wheels for easier transport of the cabin; robust wooden counter; modern design.

The process of assembling the measuring booth begins with ordering all products for the measuring booth. It is necessary to order aluminum profiles, Plexiglas, blinds, electronic and mechanical components, sensors, computer, i.e. all components that are in the cabin assembly. Also, it is necessary to draw the cabin structurally in advance, in order to know its exact dimensions. Each profile has a specific number, in order to automate the process. Each number represents a different dimension of the profile, in order to save time when ordering and assembling the cabin. Each cabin is equipped with a work surface where the entire measuring equipment will be located. The counter needs are to be dimensioned, so that mechanics don’t waste time on disassembly and additional repairs later. The counter is made of wood. Certain holders are required on the work surface such as holders for standard, temperature probe and measuring heads. In the following figure it is presented a technical drawing of the holder for the temperature probe (Fig. 4).

Fig. 4. The technical drawing of the holder for the temperature probe

The technical drawing of the measuring head holder is presented in Fig. 5.

250

N. Zuji´c and D. Dihovicni

Fig. 5. The technical drawing of the measuring head holder

3 Description of the Measurement Program for Measuring with Temperature Compensation The main program consists of a series of commands, where the lines are executed in sequence. At the beginning it is necessary to define the input/output connections, clear the screen, reset the Militron and load the reference values from the characteristic file. It is done every time during restarting of the measurement program. After initializing the system, a communication with the IR temperature sensor is started, where is entered the emission factor that calculates the exact temperature value of the aluminum product. The System.ini file is used to configure the main measurement program. It is an important file in which it is configured the functions of the program. It should be turned on or off. Functions provided by the GMS program: • • • • • • • • • • •

a path to the background image; a path to the main Program.ini file; setting the background color; enabling the purchase order selection window; enabling a selection window for selecting workers; enabling the nest selection window; enabling report output; enabling report output; enable measurement history; enabling automatic shift selection; enabling measurement output on a graph.

Dimensional Measuring System with Temperature Compensation

251

After the initialization of the measuring program, the execution of the main program follows, which is executed in a DO - LOOP infinite loop. In our case, the entire main program of the measuring system consists of 8 sets of operations or steps, which then calls certain subroutines of its own. Main software sets: • • • • • •

product temperature reading; foot switch control; product opening measurement; calling the hole recalculation feature due to temperature stretching; control of all characteristics, whether the piece is within tolerance limits; recording measurements in the measurement history.

The INPUT command is used to control the foot switch, where is checked whether the foot switch is activated or not. The INPUT statement also allows the program to wait at this step until the command we declared arrives over the communication channel. Record of the INPUT command is shown in Fig. 6.

Fig. 6. Record of the INPUT command

Meaning of command parameters: • LineNo – sequential number of steps; • Function – is called a command function; • Parameter – determining the number of the communication channel and the required command from the serial interface; • LineTrue – the number of the next step when is given the correct command; • LineFalse – the number of the next step if is not received the correct command; • Wait – indicate whether the mechanics are waiting for this command only, or whether they are checking several commands at the same time; • Comment – comment field. The diameter measurement program appears six times, the number of holes for which temperature compensation is to be measured and calculated. At the beginning of each step, it is turned on the corresponding LED on the controller, which illustrates the active measurement pin. This allows the worker to always pick up the correct measuring needle. In addition, appropriate instructions are displayed on the screen, so the worker can be guided in the correct way, to perform the measurement step.

252

N. Zuji´c and D. Dihovicni

For each new measurement step, an image is also displayed that graphically illustrates which hole is being measured. When both steps are completed, the foot switch condition is checked. When the foot switch is activated, the command to continuously read the value from the first channel of the Militron measurement interface is activated. However, the characteristics and dialing are recalculated temperature elongation characteristics. The temperature measurement software part takes care of starting the product temperature measurement and calling the temperature characteristic. After the temperature is measured and confirmed with the foot switch, the execution of the main program continues at the first step of the opening of measuring. The GETCHAR 34 command calls feature 34 from the PartChars.ini file. Number 34 talks about the feature index. When the worker presses the foot switch, the program cancels the function call and continues to the borehole measurement step. Meaning of characteristic parameters: • • • • • • • • • • • • • • • •

Name – the name of the feature; HI, LI – warning record value; HR, LR – alarm value; HG, HG – value of the exact record; Nominal - nominal value; Offset – offset value; GetData – record of the command for reading from the temperature meter; XP, IP – coordinate record of the position on the LCD screen; Precision – number of decimal places; MeterWidth – the height of the record frame on the screen; ValueHeight – the height of the numbers on the screen; ShowGage - tag for enabling graphic display - bar chart; Channel – communication channel; CalcData – a line for executing mathematical equations; CharID – Index of characteristics; BackgroundColor – background color of the characteristic frame.

Reading the probe on the first channel and calling the feature is done with the AUTOINFO command, which is shown in Fig. 7. It differs from the previous step in function and set parameters.

Fig. 7. Calling values from Militron

Meaning of AUTOINFO command parameters:

Dimensional Measuring System with Temperature Compensation

253

• LineNo – sequential number of steps; • Function – is called a command function; • Parameter – determining the number of the communication channel and the required command from the serial interface; • LineTrue – the number of the next step when is given the correct command; • LineFalse – the number of the next step if it is not received the correct command; • Wait – indicate whether mechanics are waiting for this command only, or whether they are checking several commands at the same time; • Comment – comment field. Each characteristic takes into account the calculation of temperature expansion. Thermal expansion is a phenomenon in which substances expand due to an increase in temperature. An increase in the temperature of a substance is a reflection of an increase in the internal energy of the substance. This is due to the higher energy of the building blocks of matter, such as atoms or molecules. Elongation is affected by: • temperature change (the greater the temperature change, the greater the elongation); • body size (the greater the size, length, volume of the body, the greater the increase in dimensions); • type of substance (substances expand differently due to temperature). • The temperature coefficient is an expression for two related physical quantities that show how the dimensions of a body change with temperature: • as the temperature coefficient of linear expansion (a); • as the temperature coefficient of volumetric expansion (b). In this research, the coefficient of linear expansion was used, since a piece made of aluminum alloy is measured. The volume coefficient is used for condensed substances. The temperature coefficient of linear expansion is information that tells how much the length/size/volume of a material increases if the material is heated by 1° K. Programming the GMS measuring program for one of the dimensions, is the same and procedure was repeated for each subsequent dimension:

254

N. Zuji´c and D. Dihovicni

• Buttons [Button1] Name=REPEAT_STEP_1 XPos=1676 YPos=760 Height=40 Width=200 FontSize=24 Caption=KORAK NAZAJ GotoLabel=REPEAT_STEP_1 Enabled=1 Hidden=1 [Button2] Name=REPEAT_STEP_2 XPos=1676 YPos=760 Height=40 Width=200 FontSize=24 Caption=KORAK NAZAJ GotoLabel=REPEAT_STEP_2 Enabled=1 Hidden=1

4 Workflow and Presentation of Achieved The entire request for the development of the project was submitted by the company LTH Castings d.o.o., which deals with the machining of aluminum castings. From the very idea to the final product, there was a lot of research and finding the right ways to measure temperature, because the problem was mass temperature reading, without mechanical touch. Stages of project development: • • • • • • • • •

a meeting with company leaders for given requirements; looking for information and solutions; creation and installation of a measuring system that meets the user’s requirements; programming and adjustment of the measuring system; testing and elimination of possible errors; transport and final assembly of the measuring system to the company LTH; putting the device into operation at the company LTH; informing the user about the new device; inclusion of the device in the company’s use.

In Fig. 8, it is presented appearance of the measuring cabin with a protective shutter and measuring needles. The measuring needles are placed on a stand with LED lights. At the end of use, the measuring booth is closed with a shutter to protect the components from oil and dust.

Dimensional Measuring System with Temperature Compensation

255

Fig. 8. Appearance of the measuring cabin with a protective shutter and measuring needles

In mechanical production, a measuring system with temperature compensation for the processing of aluminum castings is installed. It is placed at the end of the machining center, because it is necessary to measure each piece 100%. The use of the measuring system itself must be easy and simple, which allows burdening the workers as little as possible during work. Figure 9 shows the inner part of the measuring cabin.

256

N. Zuji´c and D. Dihovicni

Fig. 9. The inner part of the measuring cabin

5 Conclusion The task of developing a temperature compensation measuring system was a request from the client LTH d.o.o. and worked at Gazela d.o.o. from Krško, which deals with the production and implementation of measuring systems. From Gazela d.o.o. it was determined that the problem of elongation due to high temperature in the production lines and the product can be avoided by using a measuring system that contains temperature compensation. Through the realization and development of the project, the whole process should be familiar, from the problem itself to the idea and the entire solution. During the development phase, many problems were discovered, which were solved by various tests. The system also contains a wider range of electronics than power components. Communication with the computer and external peripherals is also included. In any case, during the operation and testing of the entire measuring system with temperature compensation, potential errors will appear, but they will be solved on the fly, since the system allows remote access to the computer and errors can be quickly corrected in this way. In addition to correcting errors, remote access also enables checking whether the worker or user is using the measuring system correctly. Further development is based on the improvement of the existing system and the production of measuring equipment. The focus would be on precision measuring controls, which have micron tolerance, micron measurement on counter, form, roughness and similar machines. Then there is another very important part of this production process, which consists in the assembly, programming and distribution of temperature-compensated measuring cabins.

Dimensional Measuring System with Temperature Compensation

257

.References 1. Zhaoxu, Y., Shugang, L., Hongbin, D.: Razumikhin-Nussbaum-lemma-based adaptive neural control for uncertain stochastic pure-feedback nonlinear systems with time-varying delays. Int. J. Robust Nonlinear Control 23(11), 1214–1239 (2013) 2. Dihovicni, D.J., Asonja, A., Radivojevic, N., Cvijanovic, D., Skrbic, S.: Stability issues and program support for time delay systems in state over finite time interval. Phys. A Stat. Mech. Appl. 1(538), 122815 (2020) 3. Dihoviˇcni, D., Škrbi´c, S.: Fuzzy approach to supply chain management for e-commerce store. Appl. Eng. Lett. 5(2), 62–67 (2020). https://doi.org/10.18485/aeletters.2020.5.2.4 4. Wan, J., Cai, H., Zhou, K.: Industry 4.0: enabling technologies in intelligent computing and internet of things (ICIT). In: International Conference on IEEE, pp. 135–140 (2015) 5. Shiquan, Z., Aragona, P., Shining, C.: Advances in engineering science and equipment engineering II. In: ICEESE 2016, Conference 2016, vol. 1, pp. 207–211, London (2011) 6. Dihovicni, D.J.: Decision making and fundamental matrix approach in process safety. Int. J. Comput. Intell. Syst. 6(4), 658–668 (2013) 7. Bolton, W.: Mechatronics, Electronic Control Systems in Mechanical and Electrical Engineering, 3th edn. Pearson Education Limited, England (1999) 8. Berger, H.: Automating with STEP 7 in LAD and FBD, SIEMENS, Publicis MCD Verlag, Erlangen and Munich (2003) 9. Alciatore, D.G., Histand, M.B.: Introduction to Mechatronics and Measurement Systems, 3rd edn. McGraw Hill, New York, International edition (2007) 10. Pastukhov, A., Timashov, E., Stanojevi´c, D.: Temperature conditions and diagnostics of bearings. Appl. Eng. Lett. 8(2), 45–51 (2023). https://doi.org/10.18485/aeletters.2023.8.2.1 11. Xiuzhi, L., Yaping, D.: Fault tolerant control for fixed set-point control nonlinear networked control systems. Tech. Gaz. 23(1), 35–41 (2016) 12. Xiao, L., et al.: Temperature dependent capillary rise and its effects on fabric cleaning and permeability. Therm. Sci. 27(3), 1915–1920 (2023) 13. Kovaˇcevi´c, Z., Vukašinovi´c, V., Gordi´c, D.: Piecewise techno-economic analysis of led com bulbs for street lighting. Appl. Eng. Lett. 7(2), 74–80 (2022) 14. Spasic, M., Dihovicni, D.: Protection of database in machine industry and defense of cyber attack with antivirus and firewall protection. Appl. Eng. Lett. 4(2), 66–71 (2019). https://doi. org/10.18485/aeletters.2019.4.2.4 15. Stefanini, R., Tancredi, G. P. C., Vignali, G., Monica, L.: Industry 4.0 and intelligent predictive maintenance: a survey about the advantages and constraints in the Italian context. J. Qual. Maint. Eng. 29(5), 37–40 (2023). ISSN: 1355–2511 16. Zhao, L., Ya, L., Ji-Huan, H.: Promises and challenges of fractal thermodynamics. Therm. Sci. 27(3), 1735–1740 (2023) 17. Lima, R., Barros, D.: Identification of time delay systems: a state space realization approach. IFAC Pap. Online 48(8), 254–259 (2015) 18. Janjic, N., Adamovic, Z., Nikolic, D., Asonja, A., Stojanovic, B.: Impact of diagnostics state model to the reliability of motor vehicles. J. Balkan Tribol. Assoc. 21(2), 452–463 (2015) 19. Toropov, E., Tumasov, A., Vashurin, A., Butin, D., Stepanov, E.: Hardware-in-the-loop testing of vehicle’s electronic stability control system. Appl. Eng. Lett. 8(2), 70–79 (2023). https:// doi.org/10.18485/aeletters.2023.8.2.4 20. Dihovicni, D.J., Nedic, N.: Simulation, animation and program support for a high performance pneumatic force actuator system. Math. Comput. Model. 48, 761–768. Elsevier, Washington (2008) 21. Argyll, J., Hill, L.: Concerning the amount of nitrogen gas in the tissues and removal by breathing almost pure oxygen 6I2(2), 309–321

258

N. Zuji´c and D. Dihovicni

22. Sriboon, W., Pansila, P., Khunthong, S., Sukprasong, S.: Energy audit and evaluation of specific energy consumption of lighting systems in a new building: a Kasetsart University case study. Appl. Eng. Lett. 6(4), 157–165 (2020) 23. Yong, L., Yanping, Y., Chaofeng, L., Xu, H., Xiaosong, Z.: Human responses to high air temperature, relative humidity and carbon dioxide concentration in underground refuge chamber. Build. Environ. 131, 53–62. ELSEVIER (2018) 24. Mann, B., Patel, B.: Stability of delay equations written as state space models. J. Vib. Control 16(8), 1067–1085 (2010) 25. Asonja, A., Desnica, E.: Reliability of agriculture universal joint shafts based on temperature measuring in universal joint bearing assemblies. Span. J. Agric. Res. 13(1), e02-001 (2015) 26. Ašonja, A., Adamovi´c, Ž.: Maintenance of Roller Bearings. Duga knjiga, Sremski Karlovci (2010)

3D Printing Technology: Materials, Application and Current Trends in Process Improvement Ivan Palinkas1(B)

, Eleonora Desnica1 , Jasmina Pekez1 and Milan Rackov3

, Aleksandar Rajic2

,

1 Technical Faculty “Mihajlo Pupin”, University of Novi Sad, Djure Djakovica BB,

23000 Zrenjanin, Republic of Serbia [email protected] 2 Technical College of Applied Studies, Djordja Stratimirovica 23, 23000 Zrenjanin, Republic of Serbia 3 Faculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, Republic of Serbia

Abstract. Additive manufacturing, or 3D printing, is production technology for manufacturing parts, especially parts with complex geometry or complex structures. It opens different possibilities in part mass optimization, rapid prototyping as well as sustainable production and waste minimization. It is applicable in different fields such as mechanical engineering, civil engineering, medicine, food industry and so on. Manufactured parts from additive manufacturing can be very diversive depending on used technology and applied material. On qualitative features of 3D printed part can be influenced through various segments of additive manufacturing production from 3D modeling (optimizing model for 3D printing), 3D models slicing and production parameters adjustments (from the applied slicing software), 3D printer settings and used printing technology (applied material and used printing technology such as FDM or SLA, etc.) to 3D part post process treatment. This paper describes current 3D printing technologies, materials and methods for achieving functional 3D printed parts. Keywords: 3D printing · Materials · Process improvement

1 Introduction Additive manufacturing (AM) technology is a process that can produce 3D objects through adding material, layer by layer. 3D printing is automated process for production of solid parts from digital 3D objects [1]. AM is defined by ISO/ASTM 52900:2018 standard [2], and there are currently seven AM process categories: [2–4]. • Vat Photopolymerisation – is the process where a photopolymer resin is being exposed to light (with specific wavelengths) and solidify. Technologies that are developed for this process are Stereolithography (SLA), Digital Light Processing (DLP) and Continuous Digital Light Processing (CDLP). SLA technology is using laser for curing, DLP is using projector and curing in CDLP is based on LED and oxygen. Material used for 3D printing is plastic. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 259–268, 2024. https://doi.org/10.1007/978-3-031-51494-4_22

260

I. Palinkas et al.

• Powder bed fusion (PBF) – uses a heat source to induce fusion between the powder particles layer by layer. That fusion (sintering or melting) is applied on plastic or metal powder. Technologies that are using PBF are Selective Laser Sintering (SLS), Direct Metal Laser Sintering/Selective Laser Melting (DMLS/SLM), Multi Jet Fusion (MJF) and Electron Beam Melting (EBM). SLS and DMLS/SLM are using laser, MJF is using agent and energy and EBM is using electron beam for particle fusion. SLS and MJF are applied with plastic materials, and DMLS/SLM and EBM are used with metal materials. • Material extrusion – is based on extrusion of material through nozzle. Fused deposition modeling (FDM) technology works on that principle and it is most used AM technology today. FDM works with plastic and composite materials in wire shape (filament). • Material jetting – uses liquid droplets as jetting material and then solidifies them. Technologies associated with material jetting are Polyjet - using UV light for curing and works with plastic, NanoParticle Jetting (NPJ) – using heat for curing and works with metal, and Drop On Demand (DOD) – part milled to form and works with wax. • Binder jetting (BJ) – is technology that uses printhead (industrial grade) for binding adhesive agent deposition on thin layer of powder particles. The difference between Binder jetting and Powder bed fusion is that binder jetting does not require heat. BJ technology can be applied with gypsum, sand and metal. • Direct energy deposition (DED) – is used for creation of 3D object by melting powder material as it is deposited. Technologies that are in this category are Laser Engineering Net Shape (LENS) and Electron Beam Additive Manufacturing (EBAM). LENS is using laser for particle fusion and EBAM is using electron beam. Both technologies works with metal. • Sheet lamination – uses thin sheets of materials, stacks them and laminate to make parts. Technologies based on sheet lamination are laminated Object Manufacturing (LOM) and Ultrasonic Consolidation (UC). Materials used for production are paper and composite materials. All of defined processes were initially developed for manufacturing parts from different types of polymers (sheet lamination is exception). Now, they have been upgraded and can be used, not only for creation of prototypes, but in the large-scale production and production of high-end quality parts with implementation of different materials. From 2010, there is significant increase in interest related to AM/3D printing [5]. This is shown through increased number of publications on that topic (Fig. 1), and through development of all parts of process that is necessary for part manufacturing, from design, methods for design optimization, materials improvement, print size and batch size, etc.

3D Printing Technology: Materials, Application and Current Trends

261

Fig. 1. Number of publications on AM/3D printing in respective years [5]

2 Materials Used for 3D Printing Materials used for 3D printing can be various. Mostly used materials are polymers. Also, 3D printing technologies can work with metal, composite material (either as reinforcement through plastic filament or standalone material), ceramics, biomaterials, etc. Some of the used materials regardless of the implemented 3D printed technology can be found in Table 1. Table 1. Materials used in different 3D printing technologies with characteristics and application [6] Material

Characteristics

Application/Industry

Stainless steel

High tensile strength, heat and corrosion resistant

Suitable for parts exposed to highly abrasive environments such as pump components and parts for down-hole drilling and mining equipment

Ceramic beads

Good thermal expansion and high permeability

Ceramic beads are compatible with all binders and are recommended for casting steel alloys or printing cores subject to high thermal stress conditions

Inconel alloy

Highly dense, good mechanical properties

Commonly used for gas turbine blades, seals, pressure vessels in the aerospace industry as well as steam generators in pressurized nuclear water reactors (continued)

262

I. Palinkas et al. Table 1. (continued)

Material

Characteristics

Application/Industry

Iron

Good mechanical properties and excellent wear resistant

Suitable for automotive components, machine tools, tooling, decorative hardware and iron is most widely used material for industrial applications

ABS

Tough and strong

Automotive, aerospace, medical-device

ASA

Mechanical Strength and UV stability

Functional prototyping from brackets and electrical housings to automotive prototypes and practical production parts for outdoor use under the sun

Nylon 12

Good chemical resistance, high fatigue resistance and high impact strength

Ideal material for applications that demand impact-protective components and high fatigue endurance, including antenna covers, custom production tooling, friction-fit inserts and snap fits in automotive and aerospace industries

PLA

Good tensile strength and surface quality

Ideal for model and prototypes that require aesthetic detail and environmentally-friendly for both home and office

TPU

Excellent tear and wear resistance, Exceptional flexibility (i.e. high impact strength and hardness elongation at break) and corrosion resistance to many common industrial chemicals and oils. Highly versatile material with the both rubber and plastics properties for a variety of industrial application

VeroWhitePlus

Durable, rigid and high dimensional accuracy

Suitable for a range of industries applications such as electronic housing, medical devices, work piece with complicated features (continued)

3D Printing Technology: Materials, Application and Current Trends

263

Table 1. (continued) Material

Characteristics

Application/Industry

Digital ABS

Higher heat deflection temperature Functional prototypes, injection molds, manufacturing tools, electronics enclosures, durable presentation models, engine parts and covers

Dental material

Good strength, high accuracy and durability

Three types of dental materials (VeroDent, VeroDentPlus and VeroGlaze) are approved for in-mouth placement, veneer try-ins and diagnostic wax-ups

Titanium (Ti)

Corrosion resistance, biocompatible, low thermal expansion, high strength and low density

Titanium components can be applied across a broad spectrum of applications, such as medical technology, aerospace, automotive, maritime, jewelry and design

Stainless steel

Hardenable, high resistance to Applications for stainless steel wear and tear, corrosion resistance, components are found in great hardness and high ductility automotive industry, toolmaking, maritime, medical technology, mechanical engineering

Aluminium (Al)

Good alloying properties, good process ability and electrical conductivity, low material density and light metal

Aluminium components are optimal for use in areas such as aerospace engineering, automotive industry, prototype construction, especially thin-wall components with complex geometries

Cobalt-Chrome

Biocompatible, very high hardness, corrosion resistance, high strength and high ductility

Cobalt-chrome components can be used in medical and dental technologies, high-temperature fields such as in jet engines

Nickel based alloys

Outstanding weldability, hardenable, corrosion resistance, excellent mechanical strength

Nickel based components can be used in aerospace engineering, high-temperature fields, toolmaking

Novel 3D printing material defined by [6] can be divided in following groups: • Digital materials – advanced composite material that is made of two or three photopolymers in specific microstructures and ratios. They can create functional prototype with tunable characteristics. • Smart materials – have capability to transform their geometry under the influence of external influence (basic usage is in 4D printing).

264

I. Palinkas et al.

• Ceramics materials – represents one of the challenging materials for implementation in 3D printing process because of their high melting point. • Electronic materials – material that allows 3D printing of functional electronic devices. • Biomaterials – still in development for the purpose of printing human organs. On Fig. 2 can be seen the application of one of the novel materials from the category of electronic materials as 3D printed electronic components [7].

Fig. 2. 3D printed electrical components before the removal of wax (from left: Resistor, Inductor, Capacitor and LC tank) [8]

3 3D Printing Applications Survey conducted by HUBS [9] is showing what are main applications for 3D printing in 2023 (Fig. 3) and also what are the primary industries that are using 3D printing. Main application of 3D printing is: 1. Prototyping – the creation of sample product for design and functionality assessment. Prototyping was first real use of 3D printing, and the goal is to speed up product development. It is also the method for verification of the simulation results of product design. 2. Tooling – the creation of specialized equipment for production. 3D printing, with custom part production on demand, can enhance production lines by minimizing machine downtime, by increase in production agility, and enabling custom solutions. This application has been utilized by Toyota in production through agility, faster lead time and design freedom [10]. 3. Low-volume production – 3D printing is suitable for production of limited number of units. This type of production minimizes costs associated with inventory and storage. 4. Mass customization – 3D printing enables production for individual customers without compromising high output levels. Examples of mass customization are dental aligner molds [11] and custom ergonomic computer mice based on individual users hand image [12]. 5. Serial production – creation of multiple identical products, in succession. Serial production with 3D printing is viable for small parts (several small parts can fit into single build).

3D Printing Technology: Materials, Application and Current Trends

265

Fig. 3. Survey results on question: What is your main use for 3D printing? [9]

Primary industries where 3D printing is utilized is: aerospace – it is among the first adopters. The main reason is that 3D printing allows the production of lightweight components increasing energy efficiency of aircraft. Also, enables design optimization, assembly components minimization and development acceleration with direct functional end-part production in metal or engineering-grade polymer; automotive – in this industry 3D printing processes are used for all above main applications; medical – technology is used for mass customization of implants, prosthetics, surgical guides and instruments, dental products, etc.

4 3D Printing Improvement Trends Design of functional 3D printing parts requires non-traditional approaches to product design. In order to increase chances for successful print and to shorten production time the Design for Additive Manufacturing (DfAM) framework can be implemented (Fig. 4) [13]. DfAM framework consist of elements necessary for part design in order to be manufactured on 3D printing technology. Every segment of DfAM like part customization, lightweighting, usage of internal channels or structures, functional integration, usage of designed surface structures and usage of multi-material provides characteristics improvement of final 3D printed part in design process. In part design process and improvement of its characteristics for AM, great part is application of optimization process like shape and topology optimization, and through it, usage of generative design, for achieving the lightweight functional part through 3D printing technology [14–18]. With computer hardware and software development this has become the imperative for usage in design process, for optimization the manufacturing process as a whole. Future improvement trends of 3D printing process are by implementation of Artificial Intelligence (AI). The area of AI implementation is in: Quality control for improvement of final product by process tuning. One of the examples of AI is viewed on Fig. 5 through utilization of computer vision for detection of print failures; Machine learning procedures and effects on AM; Generative design and increased usage od automatized design tools; Print optimization where AI can fine tune the printing parameters of complex printing

266

I. Palinkas et al.

jobs; and utilization of improved mass customization, especially in medical industry. [9, 19–22].

Fig. 4. Design framework - DfAM stages, actions and goals [13]

3D Printing Technology: Materials, Application and Current Trends

267

Fig. 5. Machine learning model for detecting defects [9]

5 Conclusion 3D printing technology currently today is widely used. Main reason is that is versatile, in some segments easy implemented, with development perspective. As seen, there can be used different 3D printing technologies depending on needs, and depending on required material. There are many materials developed for usage in 3D printing process that can satisfy required production demands. Trend for improvement is with the combination of computer hardware and software development, as well as development of AI system that are currently used in different areas. Due to development of technology and material, now the functional part can be manufactured using 3D printing technology, and with respect to economic production parameters, can be used side by side with conventional manufacturing technologies.

References 1. Banjanin, B., et al.: Consistency analysis of mechanical properties of elements produced by FDM additive manufacturing technology. Revista Materia 23(4) (2018). https://doi.org/10. 1590/S1517-707620180004.0584 2. International Organisation for standardization. Additive Manufacturing – General PrinciplesTerminology (ISO/ASTM Standard No. 52900) (2018) 3. What is 3D printing? https://www.hubs.com/guides/3d-printing/#chap. Accessed 15 May 2023 4. Mitrovi´c, R., et al.: Determination of optimal parameters for rapid prototyping of the involute gears. In: IOP Conference Series: Materials Science and Engineering, vol. 393, p. 0121051–012105-10 (2018). https://doi.org/10.1088/1757-899X/393/1/012105 5. Izdebska-Podsiadly, J. (ed.) Polymers for 3D printing: Methods, Properties, and Characteristics. Elsevier (2022) 6. Lee, J.Y., An, J., Chua, C.K.: Fundamentals and applications of 3D printing for novel materials. Appl. Mater. Today 7, 120–133 (2017). https://doi.org/10.1016/j.apmt.2017.02.004 7. Espera, A.H., Jr., Dizon, J.R.C., Chen, Q., Advincula, R.C.: 3D-printing and advanced manufacturing for electronics. Profress Additive Manuf. 4, 245–267 (2019). https://doi.org/10. 1007/s40964-019-00077-7 8. 3D-printing basic electronic components. https://www.thekurzweillibrary.com/3d-printingbasic-electronic-components. Accessed 16 Apr 2023

268

I. Palinkas et al.

9. 3D printing Trend Report 2023. Market insights and forecasts in additive manufacturing. HUBS a Protolabs company (2023) 10. How Toyota Factory Works with Zortrax 3D Printers. https://zortrax.com/blog/how-toyotafactory-works-with-zortrax-3d-printers/. Accessed 16 May 2023 11. 3D printing a straighter smile - 3D printing and the growth of mass customized orthodontic aligners. https://www.voxelmatters.com/3d-printing-orthodontic-aligners/. Accessed 16 May 2023 12. 3D printing game-changing custom electronics with Hubs. https://www.hubs.com/blog/for mify-case-study/. Accessed 16 May 2023 13. Vaneker, T., Bernard, A., Moroni, G., Gibson, I., Zhang, Y.: Design for additive manufacturing: framework and methodology. CIRP Ann. Manuf. Technol. 69, 578–599 (2020) 14. Petrovi´c, N., Kosti´c, N., Marjanovi´c, N., Velemir, A., Spasojevi´c, L.: Comparing truss sizing and shape optimization effects for 17 bar truss problem. Adv. Eng. Lett. 1(4), 142–147 (2022). https://doi.org/10.46793/adeletters.2022.1.4.4 15. Liu, J., et al.: Current and future trends in topology optimization for additive manufacturing. Struct. Multidisciplinary Optim. 57, 2457–2483 (2018). https://doi.org/10.1007/s00158-0181994-3 16. Kayakis, G., Kanellopoulos, I., Sotiropoulos, S., Lagaros, N.D.: Topology optimizationaided structural design: interpretation, computational aspects and 3Dprinting. Heliyon 3, e00431 (2017). https://doi.org/10.1016/j.heliyon.2017.e00431 17. Wang, L., Du, W., He, P., Yang, M.: Topology optimization and 3D printing of three-branch joints in treelike structures. J. Struct. Eng. 146(1), 04019167 (2020). https://doi.org/10.1061/ (ASCE)ST.1943-541X.0002454 18. Langelaar, M.: Topology optimization of 3D self-supporting structures for additive manufacturing. Addit. Manuf. 12, 60–70 (2016). https://doi.org/10.1016/j.addma.2016.06.010 19. Motalo, K., Nojeem, L., Lotisa, V., Embouma, M., Browndi, I.: Evaluating artificial intelligence effects on additive manufacturing by machine learning procedure. J. Basis Appl. Sci. Manag. Syst. 13(6), 3196–3205 (2023) 20. Paraskevoudis, K., Karayannis, P., Koumoulos, E.P.: Real-time 3D printing remote defect detection (stringing) with computer vision and artificial intelligence. Processes 8(11), 1464 (2020). https://doi.org/10.3390/pr8111464 21. Khan, M.I., Gopa, T., Neela, P.K.: Artificial intelligence and 3D printing technology in orthodontics: future and scope. AIMS Biophisics 9(3), 182–197 (2022). https://doi.org/10. 3934/biophy.2022016 22. Syuhada, A., Shamsudin, M.S., Omar, M.F., Ghoshal, S.K., Harun, S.W., Aziz, M.S.: Incorporating 3D metal printing with artificial intelligence in meeting aerospace demands. In: Journal of Physics: Conference Series, vol. 1892, p. 012015 (2021). https://doi.org/10.1088/ 1742-6596/1892/1/012015

Characteristics, Manufacturing, and Testing Methods of Polymer Gears: Review Ana Markovi´c, Lozica Ivanovi´c, and Blaža Stojanovi´c(B) Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia [email protected]

Abstract. During the years, numerous studies have been conducted to maximize gear performance and optimize gear mass, as vital elements of the power transmission. Because of this need, as well as overcoming the problems associated with the use of metal gears (noise, wear, high cost), gears made of polymer materials were taken into consideration. This paper aims to provide a review of the relevant literature, focusing on the parameters which affect the lifetime of polymer gears. The polymer materials that are currently used in gears manufacturing are: Polyether ether ketone (PEEK), polyoxymethylene (POM), nylon, polymethyl methacrylate (PMMA), polyamide 66 (PA66) (in this form and with additives), and others. An overview of the different manufacturing processes of polymer gears with guidelines for their application is given. Also, an account of the performance tests of polymer gears, which were conducted on devices specially developed for the needs of experiments, is given. The analysis of the test results showed that the lowest noise level is with PA gears, wear is maximum for ABS and minimum for POM gears. The future goal of the research is to evaluate the parameters of polymer gears, by comparing their performance with classic metal gears, which would enable the use of polymer gears for a wide range of engineering applications. Keywords: Polymer gears · Materials · Parameters

1 Introduction The transmission of mechanical energy from the driving machine to the working machine is carried out by transmission shafts, couplings, and gears. If the transmission of power between the driving and working machines is based on mechanical in principle, then it is about mechanical transmissions, which are the most common [1, 2]. In modern practice, the following are most often used: gear transmissions, chain transmissions, belt transmissions, toothed belt transmissions, cardan transmissions, etc. [3–9]. Of all the mechanical transmissions, thanks to their good working characteristics, gear transmissions have the greatest application. Geared power transmissions differ according to the position of the axis of the gear shaft, the shape of the teeth, the type of toothing and the material from which they are made. Modern designers’ efforts to reduce the mass and dimensions of transmissions have led to the development of new materials to produce both gears and other machine elements [10–13]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 269–282, 2024. https://doi.org/10.1007/978-3-031-51494-4_23

270

A. Markovi´c et al.

The need for numerous research, when it comes to polymer gears, arises because of their increasing application. In addition to increasing competitiveness on the market, the use of polymer materials in the production of mechanical transmissions also aims for ecologically supported production. Polymers expand the production of gears for machines and devices where the application of metal gears was not economically and functionally satisfactory. Polymer gears, in addition to being used at lower loads, are also used in other industries such as automotive, agricultural, medicine, robotics and many other industries. Plastic gears can be considered when designing the drive or driven part of a transmission in a vehicle, industrial reducers, and even the main transmission of light vehicles. This drives the frequency of switching from metal to polymer gears [14]. Considering the type of connection between macromolecules, polymers are divided into three basic groups, namely: plastomers (physical connections between macromolecules), duromers (crosslinked structure, chemical bonds) and elastomers (loosely crosslinked structure, physical and chemical). Connections, and as a subgroup, elastoplatomers that own the properties of elastomers, and the manufacturing method is characteristic of plastomer (there are only physical connections between macromolecules) [15]. The overview paper shows the tests of polymer gears according to the type of material, the method of production and certain parameters (such as: coefficient of friction, wear, geometric dimensions, speed, temperature, etc.).

2 Test Methods and Key Characteristics The aim of using an optimal design [16] polymer gear instead of a metal one is to take advantage of polymer materials and injection molding technology, and at the same time compensate for the relatively low load capacity. Guidelines for optimizing polymer gear design include increasing tooth size (higher modulus or higher pitch) to reduce root bending stress. In predicting service life, wear coefficient is a key parameter. Guidelines VDI 2736 [17] consider wear coefficients obtained from pin-on-disc tests. Using the pin-on-disc method, the following results were obtained: Tribological measurements made with a pin-on-disc machine [18] showed that the friction data for PEEK 450G varied from 0.4 to 0.6 as a function of temperature, speed, and pressure. Two different polyacetal (POM) materials (Delrin 500P and Hostaform C9091), typical of polymer gears, were used for the pin-on-disc analysis. The obtained results are as follows: friction coefficients according to the pin-on-disc test are for: Delrin 500P (1.58 × 10–4 mm3 /Nm) and Hostaform C9021 (1.26 × 10–4 mm3 /Nm) [19]. Experiments were performed on a pin-on-disk tribometer under dry sliding conditions and a constant load of 10 N and 300 rpm. The results showed that the increase in layer thickness is accompanied by a decrease in the friction force, while the best wear resistance was proved at an angle of 45°. ABS has better wear resistance than PLA [20]. In the combination of the pin-on-disc method with the numerical method, the following results were obtained: The composites were evaluated for their mechanical and tribological properties with test rig Ducom TR 20LE pin-on-disc based on the results

Characteristics, Manufacturing, and Testing Methods

271

obtained 1.5% filler composition was optimal compared to other different composition combinations. Thus, 1.5% graphene-filled acetal copolymer composites were used for the design of the gears, after which FEM analysis was conducted. The analysis concludes that acetal copolymer gears reinforced with 1.5% graphene are superior in terms of performance, weight and cost compared to metal and other polymer gears [21]. In the presented study [22], an experimental investigation of the wear behavior of coated cylindrical polymer gears made of POM was conducted. Based on the obtained experimental results, the following conclusions can be drawn: 1. Indentation tests performed on both uncoated and coated samples showed an exceedingly minor increase in hardness and indentation modulus due to the Al-coating. However, the indentation hardness and reduced modulus increased with increasing thickness of the surface coatings and were much lower than the known properties of pure aluminum. 2. SEM analysis shows that porosity appeared in the surface coating with five coating layers, which was formed during the deposition process and may have led to separation of the individual layers. 3. Adhesion analysis showed that the adhesion between the substrate (POM) and the Al coating was not high enough, which could lead to the separation of the coated surface layer at an early stage of gear operation. 4. The influence of the analyzed Al coatings on the wear of gears made of POM polymer is small and does not reduce the wear of the flanks of the gear that is connected. In addition to this method, the twin-disc test is also used: according to this test, the following conclusions were drawn: PA46+PTFE shows the best results, second one in pair surface roughness increases wear and nickel-coating of rough pair decreases abrasive wear and friction [23]. Moder et al. presented an advanced disc-on-disc (or twin disc) machine with sophisticated control technology for both dry and lubricated setups [24]. This method demonstrated the possibility of using PEEK in low slip ratio conditions, both for low and high loads, with the ability to work at high temperatures despite increased wear. Wear, friction, and temperature increase with increasing slip and load ratios. However, wear rates are significantly lower than for other polymers evaluated using the twin disc configuration. Reducing the load and slip ratio has been shown to help reduce the generated temperature and its associated effects around the tipping point and premature contact region [25]. Using the numerical method, with the help of the ABAQUS software [26], the values of the parameters were obtained, which were further compared with the analytical results according to the VDI 2736 standard [27]. Based on mathematical analysis, it can be concluded that combining numerical simulations with proper material parameters can lead to a better understanding of gear tooth deflection of polymer gears [26, 27]. The results of the numerical analysis [26] showed that it is necessary to use a suitable finite element model to obtain the proper stiffness of the gears, and therefore comparable results with the VDI 2736 standard. When choosing the type of elements, the results show: that 2D models are preferable in engineering practice because modeling time and computation time remain affordable, especially if many different configurations must be analyzed during the initial design phase. Based on mechanical properties such as tensile, compression and impact tests

272

A. Markovi´c et al.

obtained according to ASTM standards [28–31], finite element models were developed to simulate the behavior of the tested material in terms of impact, tension, and other mechanical characteristics [32]. The following results were obtained using FEA methods: Before the numerical analysis, an experimental measurement was conducted during which the gear PA 66 GFR 30 was tested at a constant load of 56, 75 Nm and 750 rpm, 1000 rpm and 1500 rpm. The failure formation of the bearing capacity of the material was also analyzed. Tooth surface temperature, tooth profile corrosion depth, tooth damage, and tooth surface were examined using scanning electron microscopy (SEM) and the corrosion behavior of the gears was analyzed [33]. When researching the behavior of a polymer gear on abrasion, the following results are reached [33]: 1. The temperature that occurs when coupling two gears of the same material (PA 66 GFR 30) is higher if the PA 66 GFR 30 gear is coupled with an AISI 8620 gear. 2. Thermal damage and cracking of heat accumulation inside the gear occurs at 1500 rpm (in conjunction with two gears made of PA 66 GFR 30 polymer material). 3. With the PA 66 GFR 30-AISI 8620 gear pair, heat spread (since AISI 8620 steel material is a good conductor) to the outside environment until thermal equilibrium was reached. Then the temperature became almost constant. 4. Operation of AISI 8620 steel gear and PA 66 GFR 30 polymer gear in conjunction was quiet. The mathematical model leads to the following results [34]: 1. The pairs that had the smallest volume were larger in width; with an increase in the width of the gears, the volume increased, and the losses decreased. 2. Profile displacements have a favorable effect on the volume and losses of the transmission pair. For all solutions, profile offsets between 0.5 and 0.55 for the smaller gear and between 0.65 and 0.7 for the larger gear were chosen (0.7 was the largest allowable offset coefficient, which was chosen based on geometric constraints). The ANSYS finite element method leads to the conclusion that the numerically obtained stress distribution values are in good agreement with the theoretical results and that the bending stress increases when the tooth surface decreases [35]. A FEM model for investigating the contact stress of coated polymer gears described the operation between a pair of gear teeth during a complete engagement cycle. An analysis of the stress in the isthmus at the point of inclination was performed. Based on the simulation results, the following conclusions were obtained [36]: 1. AGMA provided acceptable accuracy for estimating the contact stress at the tilt point under frictionless conditions. The error of the AGMA result was 1.8% compared to the simulation result. 2. Analysis of the contact stress on a pair of gear teeth during a complete cycle showed that sliding friction influences the contact stress. The root cause of this is the change in direction of sliding friction. 3. The 2 mm thick coating had a negligible effect on the contact stress during mesh cycles because the thickness was insufficient to influence the deformation behavior of the polymer gears.

Characteristics, Manufacturing, and Testing Methods

273

4. The stress distribution within the coating showed specific areas of tensile stresses caused by microslip on the contact surface. Higher COFs led to increased stresses within the coating. In the study [37], a method was developed for the analysis of the load distribution of the helical gear and the worm pair. The engagement force acting on the helical gears first increases and then decreases and agrees with the number of teeth that mesh alternately. The deformation of the helical gear varies depending on the number of meshing teeth. Thus, with this method, the load distribution and deformation of the tooth surface can be obtained. FEM analyzes lead to [38]: 1. Stress values for composite materials are less compared to cast steel gear. 2. The stress, deformation and weight of composite spur gear are less compared to cast steel gear. 3. Composite materials can be used in car gearboxes up to 1.5 kN. Experimental testing of polymer composite gears (glass fiber reinforced nylon with PTFE as internal lubricant) was carried out and two modes of failure had found one due to fatigue and the other due to wear. Fatigue can be measured directly with lifetime tests, but wear must be continuously recorded. It has been found that the performance of acetal gears is completely dependent on the load. A sudden increase in wear was observed as the torque was increased to a critical value [39]. The development of a comprehensive thermomechanical model for predicting temperature rise in thermoplastic polymer gears with any desired profile geometry in [40] leads to the following conclusions: 1. The total increase in temperature during gear operation is due to heat losses due to sliding friction (for the considered case, rolling friction accounts for only about 0.5% of losses). 2. The nominal temperature can be equated with the temperature of the root of the tooth. 3. Tolerance deviations from the ideal geometry of the tooth profile can lead to greater heat losses and a higher-than-expected temperature increase. 4. High speed thermography has proven to be an essential tool in thermal analysis and model validation as it provides an opportunity to study in detail the thermal response of a gear pair during operation. Structural polymers such as nylon, Delrin, and PEEK are used to manufacture gears through injection molding and other processing techniques. Monitoring of the condition of the polymer composite gears is conducted using the techniques of monitoring the temperature of the gear teeth and vibrations due to operation. Gear tooth surface temperature, using non-contact infrared temperature sensors, shows significant variations in gear temperature depending on torque, operating speed and material properties. The uniform wear that occurs during operation causes an increase in clearance and contributes to an increase in vibrations. Performance monitoring using an attached accelerometer predicted damage to the transmission [41]. A new test method using large teeth was developed to measure the effect of friction on tooth forces. In this method, rotation is limited; the variation of the forces is measured

274

A. Markovi´c et al.

during only one cycle of mixing a pair of teeth. The three-tooth segments are made of the investigated polymers, but the steel mating gear must be prepared in full size for balancing purposes [42]. Experimental tests on a test rig led to [43]: 1. The (average) COF of the analyzed S/CFRP gear pair under dry sliding conditions was identified as µ = 0.34. 2. Comparing the performance of the S/CFRP gear pair with the S-PEEK pair indicates a significantly superior performance of CFRP compared to PEEK in terms of service life - more than five times longer service life of the developed CFRP gears was measured compared to the reference PEEK variants. Special tests were used to examine the performance of S gears [44]. It can be argued that the profile shape of the S-gear teeth improves the contact conditions, resulting in less contact stress and heat generation. Based on temperature measurements, it was determined that the course of temperature rise in S-gears is different from that in involute gears. Scientists who have conducted a comparative study of metal, nylon and Derlin bevel gears found in transmissions believe that reducing weight will help improve performance in an automotive application at a lower cost. (If gears are produced in larger quantities) [45]. Based on the theoretical study and extensive experimental tests, it is possible to draw the following conclusions [46]: S-gears more successfully reduce noise emission than E-gears under different types of loads and at the same temperature level due to numerous advantages such as: stronger tooth root and longer convex - the concave surfaces of the connecting teeth, the lower initial pressure angle, the corresponding characteristics of the gears, the sound pressure level of S-gears (which are lower than those of E-gears). During engagement, both gears, E-gears and S-gears, exhibit slip and roll. The sound pressure level of polymer gears is proportional to torque. Test results performed with both tooth profiles (S-gears and E-gears) and at different torques revealed that S-gears are more suitable for noise reduction of gear drive systems than E-gears under normal loads and typical driving speeds.

3 Polymer Materials in Gears Manufacturing When choosing the optimal material, it is necessary to test a certain characteristic, or parameter, on adequate equipment and under the same conditions in order to determine which material gives the best results. The experiment uses a special testing device, a force sensor, a thermal camera, and a digital microscope - Keyence VHX-200. The materials evaluated are: PA6, PA6+30GF, PA66+20PTFE, PA66+30GF/15/2, POM, POM+10GF, POM+20PTFE, PPS+30CF/15. The test results [47] are summarized in several general rules for gear material selection: Application of a lubricant (PTFE) in pure polyamide (PA) causes a lower coefficient of friction. Therefore, the gears run at a lower temperature. This allows for a longer service life and/or higher transmission torque of the gear pair. The use of PTFE in combination with acetal (POM) does not improve the tribological performance of gears.

Characteristics, Manufacturing, and Testing Methods

275

The best performance of PTFE lubricants is achieved when used with fiber-reinforced PA polymer gears. More than 60% reduction in friction coefficient and 41% lower operating temperature can be achieved by using PTFE self-lubricating gear pairs [47]. Noise as one of the important parameters was monitored in relation to variations of torque, speed and combination of polymer materials. The materials that are most suitable when it comes to the noise parameter are [48]: A polymer gear pair made of PA material proved to be the quietest in operation, running a PA gear with a POM gear resulted in a very low noise level, a PA or POM gear with steel also resulted low noise. When the POM material was coupled with the same material but with grease lubrication, the noise was also low. Operation of the POM on the steel gear, but in the lubricated condition, did not result in significant differences between the noise in the unlubricated condition. The noise level is also affected by roughness. Rougher surfaces led to higher transmission noise. ANSYS analysis obtained values of deflection and VonMises stress, as well as deformation for the polymer materials that were considered. The materials considered to produce gears are nylon, polycarbonate, acetal, polyesters, polyphenylene, liquid critical polymer [49, 50]. Using the FEA method [49], the minimum stress values are reached and according to the study, analysis, results and graphics, the conclusion is reached that the best plastic material is nylon. The ANSYS analysis brought the following conclusions [50]: the size and centerline distance were the smallest in the case where one gear is made of polycarbonate and the other is made of acetal copolymer, surface stress intensity and unit load were minimal for the design in which one gear is made of nylon 6.6, and the other from acetal copolymer, nylon 6.6 is the best material for making the drive and driven gear if only the temperature rise is considered, the maximum principal and maximum shear stress are the lowest when the drive gear is made of polycarbonate, and driven by acetal copolymers. The following combinations of materials were tested on the durability test of gears on a test bench that works according to the principle of an asynchronous motor: (drive and driven): POM with protective lubricant and aramid fibers and PA66 with glass fibers and PTFE lubricant, reverse connection of the same materials, POM with protective lubricant and aramid fibers and PA66, lubricated and heat stabilized, POM with protective lubricant and aramid fibers and PA66 with glass fibers (PTFE lubricant and thermostabilized) and C45R steel and PA66 with glass fibers and PTFE lubricant [51]. The best performance is with a combination of Lubricomp KA000M (POM) and Zytel 103 HSL NC010 (PA66), both with internal lubricant and POM also with aramid fibers. The efficiency of gears of varied materials is tested on test benches designed for the study of transmission efficiency, transmission error, noise and thermal behavior of plastic gears [52]. Gear pairs are made of the following materials: POM and POM, PEEK and PEEK, PA66 and PA66, PA66+30%GF, PA66+30%GF+15% Liquid lubricant. Grease lubrication significantly increases efficiency, eliminating load and speed dependence, the effect of glass fibers in the nylon matrix is negligible, slightly increasing efficiency. Of all the materials assessed, PEEK was the most efficient and POM the least efficient. At 5 Nm of torque and for any given speed, all efficiency drops to 4 percent, which is out of range. The coefficients of friction (0.3–0.5) followed the inverse efficiency curve for any given material.

276

A. Markovi´c et al.

According to experimental tests (based on VDI Richtlinie 2736, Blatt 4 [53]) [54], the results of the tests are presented, which show significant differences between different combinations of gear materials. Zitel 103 HSL NC010, thermostabilized PA66, has been shown to serve best as a drive gear with both drive gears, namely C45R steel and Lubricomp KA000M, POM reinforced aramid. In [55] it is concluded that the friction and wear performance of nylon gear is completely different compared to acetal gears and that gear damage is root and pitch fractures instead of surface wear. With this research [56], it was concluded that the tooth surface temperature increases the most in the case of POM and the least in the case of HDPE for all different applied torques. In the case of wear, it is directly proportional to torque and inversely proportional to speed. The observed wear rate was maximum for ABS and minimum for POM. PC/ABS materials are resistant to flame, air, ultraviolet light and keep lower moisture than PA66 GFR 30% materials, their use is increasing in many industrial fields. In this study [57], it was determined that good working conditions consist of low rpm and tooth load. If the drive gear is AISI 8620, the accumulated heat on the steel gear is easily dissipated. However, when the load on the tooth increases, thermal damage and tooth breakage occurs (as well as surface melting due to the instantaneous increase in temperature). Additive materials should be added to PC/ABS materials to increase durability. A design analysis of metallic and non-metallic helical gears [58] provides data showing that the stresses induced in a nylon 66 helical gear are lower than in a metallic helical gear. The strength to weight ratio of nylon 66 gears is higher than metal gears. A gear made of polymer material provides additional advantages such as economy, light weight, self-lubrication, low noise and vibration, and is also suitable for manufacturing. Step load tests were performed at a constant speed of 1,000 rpm [59]. Significant differences in failure modes and performance were observed for the five polymer gear materials when mating them. The observed critical torques for each pair of gears are about 4.7 Nm for HDPE; 6 Nm for PC; 8 Nm for POM; 8.5 Nm for PA; and 11 Nm for PEEK. For PA and PEEK gears, progressive wear was the main failure mode observed. When comparing POM and PEEK polymer gears, it is found that the best performance is achieved with POM as the driving gear and PEEK as the driven gear, compared to POM and POM, PEEK and PEEK and PEEK and POM. Two types of commercial grade POM gears (POM-H and POM-C) were produced to compare the wear performance. Tests show a significant difference in wear performance [60]. POM-H gears have three stages of wear, i.e., run-in, linear wear and finally massive wear. Finally, massive wear occurs due to a temperature that has reached the melting point of the material, so the teeth are therefore too soft to maintain a normal connection, i.e., thermal failure [61, 62]. In contrast, POM-C gears have better thermal performance. POM-H gears have about 35% better service life on average than POM-C gears. In the study [63], topographic maps were used to evaluate the level of wear on the side surfaces of gear teeth made of polymer materials by FDM and FFF additive manufacturing techniques. We conclude that gears made of ABS M-30, Ultem 9085 and PEEK wear to different degrees. The active side of the teeth of the PEEK gear is minimally worn. This is confirmed by the topographic maps of the teeth as well as the

Characteristics, Manufacturing, and Testing Methods

277

image of the geometric structure of the surface. Tooth thickness analysis showed that the PEEK gear manufacturing process does not guarantee the stability of the gear geometry. Gear wear experiments and analysis clearly indicate that PEEK gears are the most wear resistant. Surprising results were obtained for Ultem 9085, which had the highest wear due to change in hardness. ABS M-30 gears wear less than Ultem gears [63]. In research [64], with nylon, the tensile strength is higher than the compressive strength, so material failure can be found, not only in the root of the tooth, where the tensile stress is concentrated, but also in the root of the tooth, where the pressure is concentrated. A nylon gear, which is produced by proper, controlled injection molding processes, can be used to transmit smaller forces (where steel gears are already used).

4 Manufacturing of Polymer Gears In the process of gears manufacturing, the correct choice of materials has a great influence on plastic injection molding [65]. Test procedures and sample dimensions are guided by the ASTM (American Standard for Testing and Materials) standard [28–31]: tensile strength, abrasion, fatigue. The samples were produced in a hydraulic plastic injection molding machine (Himaco model LH150–80) with a capacity of approximately 150 kg mass injection and 800 kN maximum closing force. The following materials were used: PA6, PA6 15%GF, PA6 30%GF, PA6/6, PA6/6 15%GF, PA6/6 30%GF, PA 60%GF. Processing difficulties occur with the PA 6.6 30% GF polymer, due to its lower viscosity. Finally, the influence of moisture absorption was decisive for the choice of gear material as well as the percentage addition of GF. Thus, the PAs with the best performance were PA 6 with 30% GF (driven gear) and with PA 60% GF (driven gear) [65]. For polymer gear applications, the potential of three different polymer gears (nylon, ABS and PLA) produced by the FDM process (3D printing) is investigated [66]. 1. With increasing speed, the actual coefficient of wear of polymer equipment increases. 2. The results show that the ABS gear has the maximum wear coefficient (at 600 rpm it is 0.494; at 800 rpm it is 0.867 and at 1100 rpm it is 1.363), while nylon has the minimum wear coefficient (at 600 rpm 0.229, at 800 rpm 0.529 and at 1100 rpm at 0.806). 3. Nylon gear shows the best results compared to gears made of other polymer materials. Polymer gears can be produced by injection molding and machining. According to [67], two ways to produce polymer gears are acceptable. One way is to obtain the gears by injection molding, and the other is to obtain the gears by machining. The advantages of the first method are as follows: 1. They can achieve high precision as is the case with metal gears. 2. Injection polymer gears are cheaper when it comes to machining. 3. Injection molded polymer gears are more resistant to wear and failure than metal gears. It is more pronounced with high-performance polymers. 4. These types of gears do not rust.

278

A. Markovi´c et al.

Machined gears have the following characteristics: 1. Polymer gears obtained by machine processing have lower density, and therefore reduced weight and lower inertia. 2. They require low frequency maintenance. 3. They can absorb vibrations due to the flexibility of their material. 4. They produce less noise. 5. They have a low coefficient of friction. 6. They are self-lubricating and wear-resistant when running dry. 7. They have a longer shelf life. 8. They can be used in food preparation rooms and wet environments. 9. They are resistant to corrosion. What impact does 3D printing of gears have on the performance and lifetime of the same is explained in [68]. 3D printed gears are used where complex geometries mean increased costs for conventional processes. Gears manufactured by additional processes are also suitable for prototypes or small series. Depending on the size and complexity of the gears, 3D printing is economical up to quantities of 10,000 pieces; it is the last point where injection molding becomes more economical for customers. Injection molding is especially recommended for gears of standard dimensions. But 3D printing can also support the mass production of special samples [68]. In [69], the process of gear injection (composite casting) and the process of obtaining gears by machine processing (milling of plate composites) were examined on a specific test apparatus consisting of a motor, a brake, a transmission pair, a torque meter and a camera (IR-Thermo Camera). Compared to other polymeric gear materials available on the market, gears manufactured from plate material are very suitable for applications where extremely high torques and elevated temperatures are involved. A possible area of application is actuators that are located near heat sources and do not require lubrication. The material has a weaker performance for high sliding speeds and low loads. In this range, it is superior to POM [69].

5 Conclusion Polymer materials which were researched in the review paper are: PEEK, POM, nylon, PMMA, PA66 (in this form and with additives), in addition to them, acetal, ABS, PTFE, PP are also mentioned. It has been proven that the performance of these materials, when used to make gears, depends on certain parameters. Depending on the variations of the parameters being tested, the behavior of the material also changes. The PA gear proved to be the best when it comes to noise, showing the lowest noise levels. POM gears are highly dependent on torque. The observed wear rate was maximum for ABS and minimum for POM. Grease lubrication significantly increases efficiency, eliminating load and speed dependence. When it comes to the manufacturing method, the impact of plastic injection molding, machining and 3D printing (for smaller quantities) is examined. Injection molded polymer gears are more resistant to wear and failure than metal gears. It is more pronounced

Characteristics, Manufacturing, and Testing Methods

279

with high-performance polymers. These types of gears do not rust. Due to the properties of plastic materials, injection molded plastic gears offer improved performance. Polymer gears obtained by machining have lower density, and therefore reduced weight and less inertia, require low maintenance, have the ability to absorb shocks and vibrations due to the flexibility of their material, produce less noise and have a low coefficient of friction. Tribological measurements made with a pin-on-disk machine showed thamt the friction data for PEEK 450G varied from 0.4 to 0.6 as a function of temperature, speed, and pressure. Other experiments revealed that Tg greatly affects the friction coefficients of PEEK 450G. Tolerance deviations from the ideal tooth profile geometry can lead to higher heat losses and higher than expected temperature rise. The application of polymer gears has been tested in the case of concrete mixers, worm gears, as well as in car gearboxes up to 1.5 kN.

References 1. Miki´c, D., Desnica, E., Ašonja, A., Stojanovi´c, B.: Reliability analysis of ball bearing on the crankshaft of piston compressors. J. Balk. Tribol. Assoc. 22(4-IV), 5060–5070 (2016) 2. Ašonja, A., Desnica, E., Pastukhov, A., Kuznetsov, Y., Kravchenko, I.: Methode for quick determination of the reliability level of agricultural PTO shafts. In: Blaževi´c, D., Ademovi´c, N., Bari´c, T., Cumin, J., Desnica, E. (eds.) OTO 2022. LNNS, vol. 592, pp. 143–149. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-21429-5_13 3. Blagojevic, M., et al.: Numerical and experimental analysis of the cycloid disc stress state. Tehniˇcki vjesnik 21(2), 377–382 (2014) 4. Marjanovic, N., Ivkovic, B., Blagojevic, M., Stojanovic, B.: Experimental determination of friction coefficient at gear drives. J. Balk. Tribol. Assoc. 16(4), 517–526 (2010) 5. Skuli´c, A., et al.: The impact of lubricant viscosity and materials on power losses and efficiency of worm gearbox. Tehniˇcki vjesnik 29(6), 1853–1860 (2022) 6. Stojanovic, B., Tanasijevic, S., Miloradovic, N.: Tribomechanical systems in timing belt drives. J. Balk. Tribol. Assoc. 15(4), 465–473 (2009) 7. Ivanovi´c, L., Stojanovic, B., Blagojevic, J., Bogdanovic, G., Marinkovi´c, A.: Analysis of the flow rate and the volumetric efficiency of the trochoidal pump by application of Taguchi method. Tehnicki vjesnik 24(2), 265–270 (2017) 8. Desnica, E., Asonja, A., Mikic, D., Stojanovic, B.: Reliability model of bearing assembly on an agricultural cardan shaft. J. Balk. Tribol. Assoc. 21(1), 38–48 (2015) 9. Vasi´c, M., Stojanovi´c, B., Blagojevi´c, M.: Failure analysis of idler roller bearings in belt conveyors. Eng. Fail. Anal. 117, 104898 (2020) 10. Janji´c, N., Adamovi´c, Ž, Nikoli´c, D., Ašonja, A., Stojanovi´c, B.: Impact of diagnostics state model to the reliability of motor vehicles. J. Balk. Tribol. Assoc. 21(2), 511–522 (2015) 11. Miloradovic, N., Stojanovic, B.: Tribological behavior of ZA27/10SiC/1Gr hybrid composite. J. Balk. Tribol. Assoc. 19(1), 97–105 (2013) 12. Mitrovi´c, S., et al.: Wear characteristics of hybrid composites based on Za27 alloy reinforced with silicon carbide and graphite particles. Tribol. Ind. 36(2), 204–210 (2014) 13. Stojanovi´c, B., Gajevi´c, S., Kosti´c, N., Miladinovi´c, S., Vencl, A.: Optimization of parameters that affect wear of A356/Al2O3 nanocomposites using RSM, ANN, GA and PSO methods. Ind. Lubr. Tribol. 74(3), 350–359 (2022) 14. Illenberger, C.M., Tobie, T., Stahl, K.: Recent Advances in Gearing. Springer, Cham (2022)

280

A. Markovi´c et al.

15. Šercer, M., Križan, B., Basan, R.: Design of Polymer Products. University of Zagreb, Croatia (2009) 16. Gear technology. https://www.geartechnology.com/articles/22659-optimal-polymer-gear-des ign-metal-to-plastic-conversion. Accessed 21 May 2023 17. VDI 2736. Thermoplastic Gear Wheels. VDI-Richtlinien, VDI-Verlag GmbG, Düsseldorf (2014) 18. Neil, D.: Master thesis: Modeling thermomechanical behavior of polymer gears, University of Wisconsin–Madison (2015) 19. Matkoviˇc, S., Pogaˇcnik, A., Kalin, M.: Wear-coefficient analyses for polymer-gear life-time predictions: a critical appraisal of methodologies. Wear 480–481 (2021) 20. Safak, M.T., Agca, B.V.: Wear and service life of 3D printed polymeric gears. Polymers 14(10), 2064 (2022) 21. Ramanjaneyulua, S., Sumanb, K.N.S., Phani Kumar, S., Suresh Babud, V.: Design and development of graphene reinforced acetal copolymer plastic gears and its performance evaluation. Mater. Today Proc. 4(8), 8678–8687 (2017) 22. Bonˇcina, T., Polanec, B., Zupaniˇc, F., Glodež, S.: Wear behaviour of multilayer Al-PVDcoated polymer gears. Polymers 14, 4751 (2022) 23. Moder, J., Grun, F., Summer, F., Kohlhauser, M., Wohlfahrt, M.: Application of highperformance composite polymers with steel counterparts in dry rolling/sliding contacts. Polym. Testing 66(11–12), 371–382 (2018) 24. Moder, J., Grün, F., Stoschka, M., Gódor, I.: A novel two-disc machine for high precision friction assessment. Adv. Tribol. 1–16 (2017) 25. Hoskinsa, D., Dearnb, K., Chenc, Y.K., Kukureka, S.N.: The wear of PEEK in rolling-sliding contact – simulation of polymer gear applications. Wear 309(1–2), 35–42 (2013) 26. Trobentar, B., Glodež, S., Zafošnik, B.: Deflection analysis of spur polymer gear teeth. J. Multidiscip. Eng. Sci. Technol. JMEST 2(4), 847–853 (2015) 27. VDI. VDI 2736 Blatt 2: Thermoplastic gear wheels - cylindrical gears - Calculation of the load-carrying capacity, Postfach 10 11 39, 40002 Dusseldorf (2013) 28. American Society for Testing and Materials - ASTM. D638: Standard test method for tensile properties of plastics. West Conshohocken: ASTM International, p. 16 (2010) 29. American Society for Testing and Materials - ASTM. D256: Standard test methods for determining the izod pendulum impact resistance of plastics. West Conshohocken: ASTM International, p. 20 (2010) 30. American Society for Testing and Materials - ASTM. D5963: Standard test method for rubber property - abrasion resistance (Rotary Drum Abrader). West Conshohocken: ASTM International, p. 9. Plastics Handbook (2010) 31. American Society for Testing and Materials - ASTM. D570: Standard test method for water absorption of plastics. West Conshohocken: ASTM International, p. 4 (2010) 32. Jenaris, D.S., et al.: Investigation behavior of polymer gear material. J. Mech. Continua Math. Sci. 17(5), 16–23 (2022) 33. Yakut, R., Düzcükoglu, H.: Examining the abrasion behavior of PA 66 gears in different cycles. Mater. Sci. Eng. (3–4), 1–5 (2014) ˇ 34. Miler, D., Cular, I., Hoi´c, M., Žeželj, D.: Optimization of polymer gears. 10. susret Hrvatskog društva za mehaniku. Slavonski Brod (2020) 35. Vohra, M., Vyas, K.: Comparative finite element analysis of metallic and non-metallic spur gear. J. Mech. Civil Eng. 11(3), 136–145 (2014) 36. Bae, S.-M., Seo, K.-J., Kim, D.-E.: Effect of friction on the contact stress of a coated polymer gear. Friction 8(6), 1169–1177 (2020) 37. Yonghu, Y., Hong, H., Jingning, T.: Load distribution analysis for the plastic helical gear meshing with steel worm. In: IOP Conference on Series: Earth and Environmental Science, vol. 237, p. 032080 (2019)

Characteristics, Manufacturing, and Testing Methods

281

38. Mahendran, S., Eazhil, K.M., Senthil, L.: Design and analysis of composite spur gear. IJRSI 1(4), 42–53 (2014) 39. Mao, K.: A new approach for polymer composite gear design. Wear 262(3), 432–441 (2007) 40. Cerne, B., Petkovsek, M., Duhovnik, J., Tavcar, J.: Thermo-mechanical modeling of polymer spur gears with experimental validation using high-speed infrared thermography. Mech. Mach. Theory 146(4), 1–22 (2020) 41. Senthilvelan, S., Gnanamoorthy, R.: Condition monitoring of nylon and glass filled nylon gears. In: Proceedings of the 11th National Conference on Machines and Mechanisms, New Delhi, India, pp. 278–282. Allied Publishers (2003) 42. Keresztes, R., Zsidai, L., Kalacska, G., Ando, M., Lefantd, R.: Friction of polymer/steel gear. Ann. Tech.-Sci. J. Mech. Eng. Fac. 1, 97–106 (2008) 43. Cerne, B., Bergan, Z., Sturm, R., Tavcar, J., Zorko, D.: Experimental and numerical analysis of laminated carbon fibre-reinforced polymer gears with implicit model for coefficient-offriction evaluation. J. Comput. Des. Eng. 9(1), 246–262 (2022) 44. Damijan, Z., Kulovec, S., Tavˇcar, J., Duhovnik, J.: Different teeth profile shapes of polymer gears and comparison of their performance. J. Adv. Mech. Des. Syst. Manuf. 11(6), 1–10 (2017) 45. Ingale, J.Y., Kadam, A.P., Bhosale, D.G.: Comparative study of steel, nylon 66 and delrin helical gears used in steering gearbox. IJRET: Int. J. Res. Eng. Technol. 5(6), 500–503 (2016) 46. Trobentar, B., Hriberšek, M., Kulovec, S., Glodež, S., Belšak, A.: Noise evaluation of Spolymer gears. Polymers 14(3), 438 (2022) 47. Tavˇcar, J., Grkman, G., Duhovnik, J.: Accelerated lifetime testing of reinforced polymer gears. J. Adv. Mech. Des. Syst. Manuf. 12(1), 1–13 (2018) 48. Dearn, K., Walton, D.: Acoustic emissions from polymeric gears. In: Proceedings of the World Congress on Engineering, WCE 2009, London, UK, vol. II, pp. 1832–1837 (2009) 49. Ramesh, T., Karthik, K., Ganesh, R.: Design and analysis of plastics gear for heavy duty applications by FEA method. Int. J. Mech. Eng. Technol. (IJMET) 8(7), 106–111 (2017) 50. Kumar, N., Singh, M., Kumar, A.: Design and selection of material for plastic gears. Adv. Mater. Sci. Technol. 3(1), 22–32 (2021) 51. Hlebanja, G., Hriberšek, M., Erjavec, M., Kulovec, S.: Durability investigation of plastic gears. In: Proceedings 6th International BAPT Conference, Power Transmissions 2019. MATEC Web of Conferences, Les Ulis Cedex A, France, vol. 287, pp. 1–9 (2019) 52. Walton, D., Cropper, A., Weale, D.J., Meuleman, K.: The efficiency and friction of plastic cylindrical gears. Part 1: Influence of materials. J. Eng. Tribol. 216(2), 75–78 (2002) 53. VDI 2736: Part 4. 2016 Thermoplastic gear Wheels: Determination of Strength Parameters on Gears. VDI – Richtlinien 54. Hlebanja, G., Hriberšek, M., Erjavec, M., i Kulovec, S.: Experimental determination of plastic gear durability. In: 9th International Scientific Conference - Research and Development of Mechanical Elements and Systems Kragujevac, Serbia, IOP Conference Series: Materials Science and Engineering Serbia, vol. 659, pp. 1–11 (2019) 55. Mao, K., Li, W.: Friction and wear behaviour of acetal and nylon gears. Wear 267(1), 639–645 (2009) 56. Gandhi, R., Jayawant, A., Bhalerao, A., Dandagwhal, R.: Applicability of composite polymer gear in low RPM applications – a review. Int. J. Eng. Sci. Invention (IJESI) 7(4), 36–41 (2018) 57. Yakut, R., Düzcüko˘glu, H., Demirci, M.T.: The load capacity of PC/ABS spur gears and investigation of gear damage. Arch. Mater. Sci. Eng. 40(1), 41–46 (2009) 58. Taywade, A.N., Arajpure, V.G.: Design and analysis of plastic gear and its comparison with metallic gear in the gear box of moped. Int. J. Res. Mech. Eng. 3(2), 1–6 (2015) 59. Gear technology. https://www.geartechnology.com/ext/resources/issues/1119x/polymergears.pdf. Accessed 23 May 2023

282

A. Markovi´c et al.

60. Xua, X., et al.: Wear performance of commercial polyoxymethylene copolymer and homopolymer injection moulded gears. Tribol. Ind. 43(4), 561–573 (2021) 61. Archodoulaki, V.M., Lüftl, S., Seidler, S.: Degradation behavior of polyoxymethylene: influence of different stabilizer packages. J. Appl. Polym. Sci. 105(6), 3679–3688 (2007) 62. Hertzberg, R., Skibo, M., Manson, J.: Fatigue crack propagation in polyacetal. J. Mater. Sci. 13(5), 1038–1044 (1978) 63. Pisula, J., Budzik, G., Turek, P., Cieplak, M.: An analysis of polymer gear wear in a spur gear train made using FDM and FFF methods based on tooth surface topography assessment. Polymers 13(10), 1–20 (2021) 64. Hu, Z., Hossan, M.R.: Strength evaluation and failure prediction of short carbon fiber reinforced nylon spur gears by finite element modeling. Appl. Compos. Mater. 20(3), 315–330 (2013) 65. Gasparin, A.L., Corso, L.L., Tentardini, E.K., Nunes, R.C.R., de Camargo Forte, M.M., de Oliveira, R.V.B.: Polyamide worm gear: manufacturing and performance. Mater. Res. 15(3), 483–489 (2012) 66. Mohan Harsha, K., Seetharama Rao, Y., Jagannadha Rao, D.V.N.J.: Comparison of wear behaviour of polymer spur gears using FDM process. Mater. Sci. Eng. 1168(1), 1–15 (2021) 67. IQS, Internet-site. https://www.iqsdirectory.com/articles/gear/plastic-gears.html. Accessed 15 May 2023 68. Igus, Internet-site. https://exhibitorsearch.messefrankfurt.com/images/original/mics/100 00133202101/0010431790/1637081741074_3D-printed-20gears-20brochure-20EN.pdf. Accessed 2 May 2023 69. Hans-Jörg, D., Severino, M.: Thermoplastic high performance composite gears. In: 3rd International Conference on High Performance Plastic Gears 2019, Garching/Munich, Germany, pp. 1279–1290. VDI-Berichte 2355, Germany (2019)

Energy Efficiency Enhancement in a Small Industrial Facility Ružica Kljaji´c(B) , Zorislav Kraus, Krešimir Fekete, and Predrag Mari´c Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Kneza Trpimira 2b, 31000 Osijek, Croatia [email protected]

Abstract. Small industrial utilities, connected to the distribution grid, can cause various disturbances that need to be eliminated or mitigated. Due to the complexity of the system and the large number of motors, industrial consumers may have problems with the voltage deviations caused by the excessive amount of reactive energy which is a natural consequence of their operation and construction. Since industrial consumers have to pay for both active and reactive power delivered from the grid, the goal is to reduce those costs as well as to reduce feeder loadings. Delivered energy can be reduced by installing compensation devices or by connecting photovoltaic power plant, which would produce active and reactive energy at the point of consumption. This paper analyzes the power flows and voltage conditions of a small industrial facility in three scenarios: the basic scenario, the second scenario with built-in compensation devices and in the third scenario with integrated photovoltaic power plant. For all three scenarios, power flow analysis is performed and results are presented in the form of tables and graphically. Keywords: small industrial facility · energy efficiency · reactive energy compensation · PV power plant

1 Introduction Distribution system operator (DSO) in Croatia categorizes electric energy consumers into two main categories - Households and Business. For the Business category type of network users there are number of available tariff models depending on voltage level user connects to and type of meter installed. While only White tariff model is available for the network users connected to medium, high and very high voltage network, low-voltage network users are categorized into four tariff models: Blue, White, Red and Yellow tariff model. Blue tariff model recognizes active energy (e/kWh), excessive reactive energy (e/kVarh) and metering fee (e/month) while White model further distinguishes active energy into active energy at a higher daily rate and active energy at lower daily rate (both in e/kWh). Red tariff model is intended for network users with connection power of more than 22 kW (active energy at a higher and lower daily rate, peak power, excessive reactive energy and metering fee) and Yellow model is for public lighting (active energy and metering fee). Excessive reactive energy is only charged in case of power factor (PF) being less than 0.95 or higher then 1 p.u. [1]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 283–294, 2024. https://doi.org/10.1007/978-3-031-51494-4_24

284

R. Kljaji´c et al.

Number of businesses connected to low voltage network according to [2] was 220 337 in 2021. Together they are responsible for 4 576 891 MWh or 27.2% of total 16 826 637 MWh consumption. Majority of low voltage network users are White tariff network users with 126 589 businesses responsible for consumption of 1 025 406 MWh or 6.09% of total consumption [3]. If observed trough EUROSTAT electricity consumption classes, IA class of extremely small businesses connected to low voltage defined as consumers with consumption less than 20 MWh/year and with a peak power between 5 kW and 20 kW account for 80.60% of total users and 8.79% of business consumption [4]. Due to recent market challenges and high electric energy prices, those small businesses are especially sensitive. One way to avoid unnecessary cost for White tariff network users that pay for the excessive reactive energy is to compensate it. Reactive power compensation (RPC) not only reduces the energy bill for businesses but also increases power efficiency/reduces losses and improves voltage profile by decreasing flow of reactive power in power system and thus power factor is increased. Also, successful compensation often mitigates possible power quality issues, i.e. mostly elimination of current harmonics generated by fluctuating non-linear loads [5, 6]. From a distribution network standpoint there are additional challenges. Uncompensated reactive power flow of industrial consumers causes not only considerable losses and voltage drop but can significantly reduce active power capacity of network elements [7]. Therefore, efficient reactive power compensation is important not only for small businesses, large industrial consumers but also for DSO companies. RPC can be realized using a number of different technologies where capacitor banks (CB), static Var compensators and static synchronous compensators are most common [8]. Efforts to reduce greenhouse emissions led to large scale integration of renewable energy sources (RES) into power networks. Subsidizing photovoltaic (PV) made power production on rooftops of households and businesses especially justified. Such installations are connected to grid through inverters that can also be used for compensation. CB are usually first choice for small and extra small businesses as they are most cost-efficient RPC.

2 Reactive Power Compensation in Distribution Network 2.1 Traditional Methods One of the traditionally used methods for RPC are capacitor banks which can be installed in individual, group or centralized scheme. Individual compensation is achieved by compensating reactive power of each load individually (see Fig. 1) while group compensation is compensation of a group of loads (see Fig. 2). Centralized compensation compensates reactive power of complete facility (see Fig. 3). From a perspective of control RPC can be manually operated or automatic by Power Factor Controller (PFC) [9, 10]. Furthermore, installation can be realized as single phase or three phase installation depending on type of consumers’ connection to the grid. In order to keep investment minimal, investors are prone to choose option where three phase installation in compensated with three phase capacitor banks but use PFC that monitors

Energy Efficiency Enhancement in a Small Industrial Facility

285

M Fig. 1. Individual compensation of reactive power scheme.

M

M

M

M

M

M

Fig. 2. Group compensation of reactive power scheme.

M

M

M

Fig. 3. Centralized compensation of reactive power scheme.

only one phase while compensate all three. This solution is a bit more affordable, and it is good solution for symmetrical loads. Problem with this type of RPC is when loads

286

R. Kljaji´c et al.

are asymmetrical what occurs when there are single phase loads that are not adequately divided between phases or when symmetrically divided single-phase loads do not work at the same time resulting with asymmetrical load of the grid. Such loads include air conditioners, fans, appliances with adjustable speed motors, computers, lighting, and other single-phase loads [11]. 2.2 Modern Methods Modern distribution networks contain devices such as renewable energy sources (RESs) or distributed generation (DG), battery storage, flexible loads etc. which changes the principle of distribution network operation. Traditionally passive distribution network (without power production) becomes an active one (with distributed power production) which affects the principles of RPC as well. Compensation in modern distribution network needs to be more flexible than in the traditional one and frequently used CBs are not up to the task. Photovoltaic power plants (PVs) are frequently connected to the distribution network throughout the power inverter. There is plethora of scientific papers that investigates the possibilities of PV inverters in RPC and voltage regulations [12–17]. Most of the industrially available power inverters that are used for PVs can operate in four operation modes: fixed power factor mode, volt-var control mode, volt-watt control mode and power limit mode. Most used mode is fixed power factor mode in which inverters try to maintain predefined power factor at the inverters output. If the power factor is set to one, then no RPC is possible. If the power factor is set at the value different from one, then some portion of reactive power is also injected in the network. The active and reactive power of the inverter are connected as follows:  (1) S = P2 + Q2 where: S – apparent power injection of the inverter, P – active power injection of the inverter and Q – reactive power injection of the inverter. Volt-var operation mode is most like the voltage regulation achieved by traditional synchronous generators. Power inverter tries to achieve desirable voltage value by adjusting its reactive power injection. In volt-watt operation mode, inverter tries to achieve desirable voltage value by adjusting (basically decreasing) active power injection and in the last operation mode (power limit), inverters try to limit active power injection when voltage rise above the threshold. This paper shows comparison of compensation and PV rooftop integration on power flows in a grid with small industrial facility. Industrial facility and the feeder are modeled based on a realistic data obtained by measurements. Three scenarios are observed. In baseline scenarios voltages, power infeed and currents are observed. In this scenario, no additional devices are connected. In second scenario, compensation device is integrated in industrial facility. Same parameters are observed as in previous scenario. In third scenario, rooftop PV is integrated at the location of small industrial facility and again, as in previous scenarios, power infeed, voltages and currents are observed. Paper consists of six sections. First section is introduction. Second section gives background on compensation procedure and different types of PV inverters. Third chapter is model and scenario description. Results for all scenarios are given in section four and discussion continues in fifth section. Conclusion is given in sixth section.

Energy Efficiency Enhancement in a Small Industrial Facility

287

3 Small Industrial Facility Small industrial facilities are often localized on a distribution feeder that are characterized by a un-balanced loadings. Distribution feeder is modeled based on a realistic cable and load characteristics and shown in Fig. 4. There are total of 14 loads, with one being small industrial facility. Consumers are grouped as 10 loads connected on a bus 2, small industrial facility on a bus 3 and 3 consumers connected to a bus 4. Cable characteristic is given in Table 1. Total cable length is 1,44 km. Table 1. Cable characteristic

Type1

R1, /km

X1, /km

Rn, /km

Xn /km

C1 µF/km

Cn, µF/km

0.2542

0.080424

0.1271

0.04

0.84

0.84

Load characteristics are given in Table 2. Table 2. Load characteristic Phase A P [kW]

Power factor (ind)

Phase B P [kW]

Power factor (ind)

Phase C P [kW]

Power factor (ind)

Consumer

2.8

0.99

1.57

0.96

1.47

0.920

Facility

5.874

0.803

5.027

0.792

3.170

0.744

Three scenarios are analyzed: Scenario 1 – small industrial facility is modeled as a constant load with unbalanced loading of each phase. Power flow analysis is performed and bus voltages and profile are observed as well as total losses in the modeled grid. Scenario 2 – industrial facility is compensated with fixed compensating devices. A shunt capacitor with total power of 12 kVAr is added in parallel with fixed load and again, as in Scenario 2, voltage profile, bus voltages and losses are observed. Scenario 3 – on the rooftop of the small industrial facility an PVPP is installed. PVPP supplies facility with both active and reactive power (12 kW, power factor 0.95). Same parameters are observed as in previous scenarios and their results are compared with basic scenario.

4 Results Scenario 1 In basic Scenario, due to high installed power, voltage at the end of the feeder is equal to 0.90 p.u. By Croatian Grid code, 10% deviation is allowed, which means that

288

R. Kljaji´c et al.

Fig. 4. Grid model

Table 3. Bus voltages, Scenario 1 Phase A [p.u.]

Phase B [p.u.]

Phase C [p.u.]

Bus 1

1.00577

1.00577

1.00577

Bus 2

1,00

1.00

1,00

Bus 3

0.92

0.91

0.93

Bus 4

0.91

0.90

0.93

voltage is on the margin of allowed deviation. Resulting voltages for each bus and phase is given in Table 3 and the values are expressed in per unit (p.u.). As it can be seen, voltages are unbalanced as a result of unbalanced loads in the system. Further load increase will result with voltage decrease bellow allowed margin. Voltage

Energy Efficiency Enhancement in a Small Industrial Facility

289

profile is also shown in Fig. 5. At the end of the feeder, unbalance is more highlighted than on the feeders beginning.

Fig. 5. Voltage profile, Scenario 1

Total infeed from the grid is 94.445 kW and 31.868 kvar. Total grid losses are 4.715 kW and 1.543 kvar. Current in line supplying Bus 3 is equal to 55.5 A. Scenario 2 In second scenario, compensation devices are added to small industrial facility. Total integrated power is equal to 12 kvar. After compensation is made, voltage on Bus 3 is increased and total lose sin system are reduced. Voltages are given in Table 4. Table 4. Bus voltages, Scenario 2 Phase A [p.u.]

Phase B [p.u.]

Phase C [p.u.]

Bus 1

1.00577

1.00577

1.00577

Bus 2

1,00

1.00

1,00

Bus 3

0.93

0.92

0.94

Bus 4

0.92

0.91

0.93

Voltage at the end of the feeder is slightly increased, and the increase is mostly notable on phases A and B. Voltage profile is shown in Fig. 6. In this Scenario, external infeed is equal to 93.909 kW and 21.312 kvar. Reactive power is reduced as well as the grid losses. With 4.178 kW and 1.336 kvar, which is decrease of 11.39% in active power losses and 13.4% in reactive losses. Current in line between busses 2 and 3 is equal to 50 A. With compensation devices, total loading is decreased, improving both electrical and thermal status of the feeder.

290

R. Kljaji´c et al.

Fig. 6. Voltage profile, Scenario 2

Scenario 3 In Scenario 3, PV power plant is implemented at the rooftop of small industrial facility. Based on a installed active power (approximately 12 kW), installed PV power will be 12 kW. Inverter has power factor 0.95, which means that this PVPP has ability to control reactive power and voltage at the consumer location. In this scenario, voltage is much higher than in previous scenarios. On bus 4 voltage in phase A is equal to 0.94 and in phases B and C 0.93 and 0.96 respectively. Voltage profile is shown in Fig. 7. Voltages are still unbalanced but far from allowed margin (Table 5). Table 5. Bus voltages, Scenario 3 Phase A [p.u.]

Phase B [p.u.]

Phase C [p.u.]

Bus 1

1.00577

1.00577

1.00577

Bus 2

1,00

1.00

1,00

Bus 3

0.95

0.94

0.96

Bus 4

0.94

0.93

0.96

PV produces 12 kW and 3.9 kvar and due this additional production, external infeed is reduced to 80.677 kW and 27.213 kvar. Grid losses are 2.95 kW and 0.834 kvar. This means that active power decreased in 37.4% and reactive power in 46%. Current in line connecting buses 2 and 3 is 34.8 A which is further improvement.

Energy Efficiency Enhancement in a Small Industrial Facility

291

Fig. 7. Voltage profile, Scenario 3

5 Discussion Adding compensation devices or PV at a small industrial facility will positively effect on voltages in the whole grid. In Fig. 8, voltages on bus with industrial facility are shown. As it can be seen, both methods have advantages but PV integration will cause significant voltage increase. This will also reflect on last bus in the system. Adding compensating devices improves voltage in 1.1% while adding PV improves it in 3.2%. Adding both options would result in future increase which may lead to voltage higher than allowed in some grids with lower loads. Adding compensation devices and PV will also have influence on energy sent to grid. Since part of reactive power is provided on place of consumption, less active power will be sent from the external grid into the feeder. For consumer, this means that he will not take excessive energy from the grid but it will be locally available. On the other hand, integrating rooftop PV provides both types of energy right at the place of consumption and thus reducing infeed from the grid (Fig. 9). Another improvement of suggested measures is reducing current in feeder. Figure 10 shows that adding compensation device reduces current in 10% in comparison to baseline scenario. Integrating PV reduces it over 37%.

R. Kljaji´c et al.

Voltage [p.u.]

0.98 0.96 0.94 0.92 0.9 0.88 Bus3 Phase A Scenario1

Bus3 Phase B Scenario2

Bus3 Phase C

Scenario3

35 30 25 20 15 10 5 0

AcƟve power [kW]

100 95 90 85 80 75 70 Scenario1

Scenario2 Infeed P

Scenario3 Infeed Q

Fig. 9. Power infeed from the external grid

60 50 40 30 20 10 0 Scenario1

Scenario2 Current [A]

Fig. 10. Feeder current comparison

Scenario3

ReacƟve power [kvar]

Fig. 8. Comparison on voltages on Bus 3

Current [A]

292

Energy Efficiency Enhancement in a Small Industrial Facility

293

6 Conclusion Small industrial utilities have significant impact on economics and social aspect of community in their vicinity. In order to constraint and reduce costs of their operation, various methods of energy savings can be applied. Integrating compensation devices is one technical solution which has both positive impact on facility but also on the nearby grid. Adding compensation reduces reactive energy that is taken from the grid but from the point of the grid it means lower infeed of reactive power to the whole grid. As a result, voltages are increasing and feeder loading decreases which results in lower losses. Further improvement is integrating rooftop PV, which provides most of the energy for the consumer, leading to the additional savings. This makes significant reduction in both active and reactive infeed, even lower cable ratings than with compensation devices. Since maximum production and consumption do not necessarily coincide, load shifting or adjusting work schedule or combination of both compensation types should be analyzed. Also, in this paper no economical aspect is given as the focus was on the methods of reducing losses and compensation of reactive energy. In future work, this aspect will be covered by determining the financial benefits and return-of-investment period as well as analyzing the combination of both compensating devices and PV. Acknowledgment. This work was supported by Croatian Science Foundation under the project “Prosumer-rich distribution power network” (project number: UIP-2020-02-5796).

References 1. HEP – Operator distribucijskog sustava d.o.o. (HEP ODS Homepage. https://www.hep.hr/ ods/korisnici/poduzetnistvo/157. Accessed 04 July 2023 2. Energetski institut Hrvoje Požar (EIHP) Homepage. https://eihp.hr/wp-content/uploads/2023/ 01/Energija%20u%20HR%202021_WEB_LR.pdf. Accessed 05 July 2023 3. HERA Homepage, Godišnje izvješ´ce za 2021. godinu. https://www.hera.hr/hr/docs/HERA_i zvjesce_2021.pdf. Accessed 05 July 2023 4. EUROSTAT - Statistical office of the European Union Homepage. https://ec.europa.eu/eur ostat. Accessed 05 July 2023 5. Dixon, J., Moran, L., Rodriguez, J., Domke, R.: Reactive power compensation technologies: state-of-the-art review. Proc. IEEE 93(12), 2144–2164 (2005) 6. Jianguo, Z., Qiuye, S., Huaguang, Z., Yan, Z.: Load balancing and reactive power compensation based on capacitor banks shunt compensation in low voltage distribution networks. In: Proceedings of the 31st Chinese Control Conference, Hefei, China, pp. 6681–6686 (2012) 7. Jurák, V., Bukvišová, Z., Ptáˇcek, M., Topolánek, D., Orságová, J.: Compensation of reactive power in LV network and its impact on reactive power flow through distribution grid. In: Proceedings of the 2020 21st International Scientific Conference on Electric Power Engineering (EPE), Prague, Czech Republic, pp. 1–5 (2020) 8. Miron, A., Cziker, A.C., Ungureanu, S, ., Beleiu, H.G., D˘arab, C.P.: Reactive power compensation at industrial consumers: Romanian study case. In: Proceedings of the 2022 International Conference and Exposition on Electrical and Power Engineering (EPE), Iasi, Romania, pp. 101–106 (2022) 9. Kuzle, I.: Kompenzacija jalove snage. Fakultet elektrotehnike i raˇcunarstva Zagreb, Zagreb (2010)

294

R. Kljaji´c et al.

10. Crushtymks Homepage. https://crushtymks.com/hr/energy-and-power/528-installation-pro tection-and-connection-of-capacitor-banks.html#google_vignette. Accessed 09 July 2023 11. Patidar, R.D., Singh, S.P.: Harmonic, reactive and neutral currents compensation and load balancing in 3H4W distribution systems. In: 2009 International Conference on Computer Engineering and Technology, Singapore, pp. 512–516 (2009) 12. Yung, Y., Han, C., Lee, D., Song, S., Jang, G.: Adaptive volt–var control in smart PV inverter for mitigating voltage unbalance at PCC using multiagent deep reinforcement learning. Appl. Sci. 11, 8979 (2021) 13. Dall’Anese, E., Dhople, S.V., Johnson, B.B., Giannakis, G.B.: Decentralized optimal dispatch of photovoltaic inverters in residential distribution systems. IEEE Trans. Energy Convers. 29, 957–967 (2014) 14. Dubravac, M., Fekete, K., Topi´c, D., Barukˇci´c, M.: Voltage optimization in PV-rich distribution networks—a review. Appl. Sci. 12, 12426 (2022) 15. Lee, H., Tae, D., Rho, D.: Voltage control method of micro hydropower generators for voltage stabilization in distribution feeder with renewable energy sources. Tech. Gazette 27(5), 1557– 1562 (2020) 16. Astapov, V., Trashchenkov, S., Gonzalez-Longatt, F., Topi´c, D.: Performance assessment of TSO–DSO using volt-var control at smart-inverters. Int. J. Electr. Comput. Eng. Syst. 13(1), 48–61 (2022) 17. He, P., Zhang, X., Li, C., Yun, L., Yang, H., Yan, Y.: Smooth regulation of DC voltage in VSCMTDC systems based on optimal adaptive droop control. Tech. Gazette 30(4), 1234–1240 (2023)

Comparative Study of Single-Input and Dual-Input PSS in Multi-machine System Tomislav Košorog, Muharem Mehmedovi´c, Predrag Mari´c, and Ružica Kljaji´c(B) Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Kneza Trpimira 2B, Osijek, Croatia [email protected]

Abstract. To meet growing demand for electricity, power systems are operating near their operating limits. Due to technical limitations, the transmission lines are close to their transmission capacity and with different dynamic characteristics interaction can occur between the parts of a single power system or between interconnected power systems. Such interactions can result in low-frequency oscillations that further limit the transmission capacity of the system. Low-frequency oscillations can be reduced by increasing the damping torque of the synchronous generators. The most efficient way to increase damping torque is to implement power system stabilizers in the generator excitation circuit. In this paper, singleinput and dual-input PSS are implemented in IEEE 14-Bus test system and their influence on the small-signal stability is compared. PSS location is determined by participation analysis, while tuning of the PSS is performed using the pole placement method. For the different tuning parameters, the system performance was analyzed in both time-domain and S-domain for single-input and dual-input PSSs implementations. Keywords: Single-input PSS · Dual-input PSS · Participation analysis · Small-signal stability

1 Introduction Bulk power systems have been equipped with significant inertial contributions of the synchronous generators that may reduce the negative effects of the disturbances. Synchronous machine is one of the main elements in power system and due to its complexity and construction features has a great impact on the system dynamics. The output power and the voltage on the generator’s terminals can be controlled using the excitation system performing protective and control functions necessary for stable operation. Natural damping of the synchronous generator is rather low and can be increased implementing power system stabilizers (PSS) within the excitation system [1]. Power system stabilizers enhance system stability and therefore are widely implemented on large generators and in multi-machine systems. In such systems, maintaining system parameters is crucial especially if some sensitive loads or elements have to be kept in strict operating margins [2]. For example, generators in nuclear power plants have very strict operating and © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 295–310, 2024. https://doi.org/10.1007/978-3-031-51494-4_25

296

T. Košorog et al.

maintenance rules in order to maintain and improve the stability [3]. In paper [3], PSS implementation in the system with nuclear power plant resulted in more efficient low frequency oscillation damping i.e., better oscillatory stability. Beside the bulk power systems, PSS is also efficient in modern power systems, where the system inertia is reduced due to high renewable energy sources penetration [4]. PSS primary function is enhancing small-signal stability, but with proper tuning, PSS can also be used in improving transient stability, as suggested in [5–7]. Authors [8] base their PSS tuning approach and adjustment on a fault condition what has resulted in the better oscillation damping. There are four main types of PSSs: single input PSS (type PSS1A), dual-input PSS (PSS2A), dual input single-band PSS (PSS3B) and dual-input multi band PSS (PSS4B) [9]. Single input PSS uses generator speed, frequency or active power as input signal while dual-input PSS uses combination of two signals. The most common combinations are speed for single input and speed-active power for dual-input PSS [10]. This paper compares the impact of single-input and dual-input PSS on system performance using a standard IEEE 14-bus system as an example. Small signal and transient stability are analyzed with respect to a small disturbance. Active power, rotor angle, and generator speed are observed in the time-domain, while the root locus and characteristic eigenvalues are observed in the S-domain. The location of the PSS is determined by analyzing the participation of all synchronous generators in the dominant oscillatory mode in the system, while the PSS parameters are tuned using standard recommendations based on the dominant oscillatory mode properties in the S-plane. System performance indices are compared with and without PSS implementation and shown graphically as well. Paper consists of 4 sections with Introduction as first section. In the second section, theoretical background for participation analysis and PSS tuning is given. Third section presents an overview of used model(s), scenario’s descriptions, results, and discussions. Last chapter contains conclusions.

2 Eigenvalues and Participation Analysis Small-signal stability is the ability of a system to maintain stable work after being submitted to a small disturbance and it is defined for a linearized system around operating point. Mathematical background for linearization is given in [11] and [12] and the same literature is used for describing eigenvalues and participation analysis: Matrix n-dimensional matrix A is described as in Eqs. (1) and (2) Avk = vk λk

(1)

where vk – right eigenvector of matrix A λk – eigenvalue of matrix A and wk A = λk wk where wk – left eigenvector of matrix A

(2)

Comparative Study of Single-Input and Dual-Input PSS

297

Eigenvalue is defined as λk = α ± jβ

(3)

where α - damping constant β - frequency of the damped oscillation Eigenvalues presents relationship between system modes and state variables as eigenvectors are linked with physical units. Participation of each variable in mode λk is determined by relative unit called participation factor given with Eq. (4) pki = v ki · wik

(4)

where pki - participation of the state variable x k in mode λi . vki - k-th entry of the right eigenvector vi . wik - k-th entry of the left eigenvector wi . Participation factors can be shown in form of matrix as: ⎤ ⎡ ⎤ v1k wk1 p1k ⎢ p2k ⎥ ⎢ v2k wk2 ⎥ ⎢ ⎥ ⎢ ⎥ pk = ⎢ . ⎥ = ⎢ . ⎥ ⎣ .. ⎦ ⎣ .. ⎦ ⎡

pnk

(5)

vnk wkn

And simplified as  P = p1 p2 · · · pn

(6)

The participation factors are normalized, and their sum is equal to 1, which means that the sum of all participations in the observed mode must be equal to 1. The location of the PSS is determined according to the participation analysis for the dominant oscillatory mode of the system, i.e., according to the contribution of each active component to the generation of the oscillations characterizing the dominant oscillatory mode. The synchronous generator in the system with the largest participation can be considered optimal for PSS implementation. PSS parameters are defined based on a standard recommendation for tuning of excitation systems [11, 13–17] as well as on an experiential knowledge. In this paper, pole placement method will be used for tuning both types of PSS. Tuning of the Single-Input Power System Stabilizer Single input stabilizer, commonly known as PSS1A, consists of wash-out filter, two lead-lag filters and gain, as shown in Fig. 1. Optimal operation of single input PSS is achieved optimizing time constants T 1 , T 2 , T 3 , T 4 and gain K. Washout constant is set to 10 s in order to avoid influence of normal signal deviations [12–14]. Another simplification is to consider time constants T 1 = T 3 and T 2 = T 4 . In this way, optimization problem is reduced on finding two parameters, namely time constant T 1 = T 3 and PSS gain constant - K [12–14].

298

T. Košorog et al. Wash-out filter Tw

Lead-lag filter T1, T2

Lead-lag filter T3, T4

Gain K

Fig. 1. Single-input PSS structure

PSS time constant is determined with iterative variations of its initial value recommended in [12–14] - T1initial = T 1 = T 3 = 0,2s observing the influence on the dominant mode damping while the PSS gain remains the same. The iteration process ends with value of the time constant that corresponds to the maximal damping of the dominant oscillatory mode and that value is considered as T 1 = T 3 . The same iterative procedure is performed for the PSS gain constant-K determination; the value-K has been varying from initial value (recommended in [11]) with fixed time constant until the maximal damping of the dominant oscillatory mode has not been achieved and that value is considered as an optimal PSS gain value. Tuning of the Dual-Input Power System Stabilizer Dual-input stabilizer (Fig. 2) uses two signals as input and each signal needs to pass wash-out filters and transducers in order to compensate for signal delay. Gain K s3 is for fine tuning of weight sum of input signals and it is mostly set according to standard recommendations [13]. For dual-input stabilizers there are also some simplifications to obtain optimal value of parameters. Simplifications are as follow [13]: – – – –

Wash-out filter constants are the same T w1 = T w2 = T w3 Transducer constant T 7 = T w1 = T w2 = T w3 time constants T 1 = T 3 and T 2 = T 4 T7 transducer gain Ks2 = 2H , where H – inertial constant of the generator

Wash-out filter Tw1

Wash-out filter Tw3 Lead-lag filter T1, T2

Wash-out filter Tw2

Transducer T6

Wash-out filter Tw4 Lead-lag filter T3, T4

Transducer T7, Ks2

Sum

Gain Ks3

Ramp filter T8, T9, M, N

Sum

Gain Ks4

Gain Ks1

Fig. 2. Dual-input PSS structure

The tuning procedure to determine the optimum value of time constant and gain is the same for each stabilizer type.

3 Case Study Proposed small-signal stability enhancement was examined on a standard IEEE 14 bus system (Fig. 3). This system has three voltage levels (69 kV, 18 kV and 13.8 kV), 16 transmission lines, four transformers, three static compensators and 11 loads. There are

Comparative Study of Single-Input and Dual-Input PSS

299

five generators and each generator is equipped with AVR. Small-signal and transient stability is analyzed for a disturbance duration of 30 ms and then the steady-state is restored. The location of PSS integration is chosen according to the participation analysis, i.e., the ability of the system state variables to affect the observed eigenvalue. Once the participation analysis is complete, the PSS is integrated into a generator that defines the stability margin, i.e., the dominant oscillatory mode and is tuned according to the procedure described in the previous chapter.

Fig. 3. IEEE 14-bus system single-line diagram in DIgSILENT PowerFactory simulation interface

300

T. Košorog et al.

Two simulation scenarios with two different types of PSS are observed. In the Scenario 1, a single-input PSS (PSS1A) is implemented with the indices observed in the time domain (rotor angle, generator speed, and active power) and the oscillatory modes are highlighted in the system S-plane. Scenario 2 is characterized by the same analysis but with the implementation of the dual-input PSS (PSS2A). Since the system consists of five synchronous generators, five characteristic oscillatory modes are shown in Fig. 4. 13.000 Imaginary part [rad/s] 7.8000

2.6000

-5.0000

-4.4000

-2.8000

-1.2000

0.4000 Real part 2.0000 [rad/s] -2.6000

-7.8000

-13.000 Stable Eigenvalues Unstable Eigenvalues

Fig. 4. Root locus for IEEE 14-bus system without PSS

Dominant oscillatory mode represents the largest real value in the S-plane and defines dynamic stability margin as well. Results of the participation analysis are shown in Fig. 5 and indicate that Generator 1 defines system stability margin. State variable used for the participation analysis is the generator speed; each grid component contributing with more than 0.002 is shown in Fig. 5.

Comparative Study of Single-Input and Dual-Input PSS

301

1.2

1

0.97

0.8

0.6

0.4

0.2 0.013

0.008

0.005

0.01

0.051

0 Generator 1 Generator 2 Generator 3 Generator 4 Generator 5

Grid

Fig. 5. Participations for the dominant oscillatory mode (mode 1)

As shown in Fig. 5, Generator 1 has the most influence on the state variables for the dominant oscillatory mode and indicates the location for PSS implementation. The tuned PSS parameters are listed in Table 1. Table 1. PSS parameters Parameter

PSS1A

Parameter

PSS2A

Washout time constant T w [s]

10

Washout time constant T w1 , T w2 , T w3 [s]

2

Lead-lag time constant T 1 , T 2 [s]

0.44

Lead-lag time constant T s1 , T s3 [s] 0.3

Lead-lag time constant T 3 , T 4 [s]

0.02

Lead-lag time constant T s2 , T s4 [s] 0.02

Gain K

100

2nd signal transducer time constant 2 T 7 [s] PSS gain K s1

25

2nd signal transducer factor Ks2

0.194

Washout coupling factor Ks3

1

Ramp Tracking N

1

Ramp Tracking M

5

Inertial constant H [s]

5.148

302

T. Košorog et al.

PSS influence on power system stability is tested performing a small system disturbance with duration of 30 ms on the Bus 4. Results Three specific variables are observed in the time domain: generator speed, rotor angle, and active power. The generator speed and rotor angle are selected because of their relationship with the synchronizing and damping torque, while the active power is one of the most important variables for generator operation. Scenario 1 In Scenario 1, PSS1A is implemented on Generator 1 with the parameters given in Table 1. Figure 6 shows the resulting active power before and after the disturbance occurrence. The blue curve represents active power without PSS implementation. As it can be seen, without additional damping torque provided by PSS, the active power oscillates for almost 10 s after the disturbance clearance with a maximum overshoot between 442 MW and 458 MW. With the PSS implementation (green-curve), the maximum overshoot of the active power is much lower (443 MW) with the settling time significantly shorter as well.

Fig. 6. Active power-without PSS-blue curve; with PSS1A-green curve

The initial rotor angle changes rapidly after the perturbation, and without PSS implementation it oscillates much more (Fig. 7 - blue curve). With implemented PSS, the undershoot of the rotor angle after the same disturbance is much smaller (Fig. 7 - green curve), and there are no further oscillations.

Comparative Study of Single-Input and Dual-Input PSS

303

Fig. 7. Rotor angle without PSS-blue curve; with PSS1A-green curve

The third observed variable is the generator speed - Fig. 8. After the disturbance occurs, the speed starts to change rapidly, and after the disturbance is removed, it reaches a new steady-state value that is slightly lower than the value before the disturbance. PSS reduces these oscillations to a minimum and, as with the rotor angle, allows a smooth transition to the new steady-state value.

Fig. 8. Generator speed without PSS-blue curve; with PSS1A-green curve

304

T. Košorog et al.

Scenario 2 As in the previous scenario, the same three variables are observed. When a PSS2A is implemented, the dynamic behavior of the system improves, as can be seen from the behavior of the active power (Fig. 9). Compared to the behavior without PSS, the oscillations and overshoot are much lower. The steady state is reached in a much shorter time.

Fig. 9. Active power- without PSS-blue curve; with PSS2A-red curve

Recovery of the rotor angle also occurs with fewer oscillations and is much smoother than without PSS (Fig. 10). Compared to the base case, it takes longer to reach the steadystate value, but since no oscillations occur during this process, it is clear that the use of PSS is recommended. Dual-input PSS also help to reduce speed oscillations of the generator. After the disturbance, the speed oscillations are much smaller and have only two overshoots before reaching a new steady-state (Fig. 11). It is important to emphasize that reaching the steady-state generator speed takes the same time, but during this process there are no additional oscillations as in the case without PSS.

Comparative Study of Single-Input and Dual-Input PSS

305

Fig. 10. Rotor angle- without PSS-blue curve; with PSS2A-red curve PSS2A

Fig. 11. Generator speed- without PSS-blue curve; with PSS2A-red curve

4 Discussion In the previous Scenarios, a comparison was made between the base case-without PSS implementation and implementation of PSS1A and PSS2A. It is clear that the introduction of PSS into the excitation circuit allows for faster fault recovery and achievement of steady state in much less time. Figures 12, 13 and 14 compare system response with implantation of different power system stabilizers. Figure 12 shows the active power of generator 1. The red curve is the characteristic obtained with the implemented PSS2A;

306

T. Košorog et al.

the initial response to a disturbance is the same, but the second overshoot is much lower than with PSS1A. Also, the steady state is reached much faster than with PSS1A.

Fig. 12. PSS1A and PSS2A comparison for active power response

The same considerations apply to the rotor angle response. Figure 13 shows the comparison for both types of PSSs. PSS2A performed lower deviation from the initial value and smoother characteristics of the transient.

Fig. 13. PSS1A and PSS2A comparison for rotor angle response

Comparative Study of Single-Input and Dual-Input PSS

307

Figure 14 shows the comparison of the generator speed for both stabilizers. The advantages of PSS2A are again evident in the smaller deviations from the initial value.

Fig. 14. PSS1A and PSS2A comparison for generator speed response

Implementation of PSS in a single generator of the system improves its dynamic behavior when small disturbance occurs. In the S-domain, Table 2, can be seen that the stability margin is shifted to the left side of the plane, resulting in a more stable system. The implementation and good tuning of PSS resulted in improved small signal stability as well as improved transient stability. The stability margin is increased as the dominant pole was shifted to the left side of the plane. Table 3 shows the comparison between oscillatory modes in the system without implementing the PSS, and with implementation of PSS1A and PSS2A. Table 2. Comparison of eigenvalues for all scenarios Mode

1

2

3

4

5

Eigenvalue without PSS −0.586104 −2.285635 −3.058144 −3.961201 ±8.564824j ±9.398019j ±10.93543j ±10.9522j

−4.146453 ±11.79519j

Damped frequency [Hz] 1.3631

1.8772

1.4957

1.7404

1.7430

Eigenvalue with PSS1A −2.037008 −2.302674 −3.060683 ±5.610014j ±9.366904j ±10.9357j

−3.960284 −4.146632 ±10.94369j ±11.79519j

Damped frequency [Hz] 0.8928

1.7417

1.4907

1.7404

1.8772 (continued)

308

T. Košorog et al. Table 2. (continued)

Mode

1

2

3

4

5

Eigenvalue with PSS2A −2.301541 ±9.36508j

−2.569666 −3.060829 ±5.196055j ±10.9356j

−3.959827 −4.146642 ±10.94324j ±11.79518j

Damped frequency [Hz] 1.4904

0.82697

1.7416

1.7404

1.8772

The benefits of PSS implementation are mainly in the improvement of small-signal stability and the improvement of transient stability. Compared to a case without PSS, it is clear that fault recovery is shorter and fewer oscillations occur. Oscillations can be permanent as the transmission system operates near its stability limits. If another disturbance occurs during these oscillations, e.g., during maximum transmission power, the grid could become unstable without being able to reach a new steady state.

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

0.9

0.105 0.002

0.061

0.043

Generator Generator Generator Generator Generator 1 2 3 4 5

0.009 Grid

Fig. 15. Participations for the dominant oscillatory mode (mode 1) after implementation of the PSS in the excitation circuit of the Generator 1

The implementation of PSS will not solve all stability problems in the system but will improve the existing condition. In the IEEE 14-Bus system, there are 5 generators and implementing PSS on one generator will have a positive effect on the overall system dynamics. The original dominant pole defined by the dynamics of Generator 1 will be shifted to the left side of the S-plane and the new dominant pole will be the one defining the stability margin. Performing a participation analysis for this new dominant pole (Fig. 15), it becomes clear that the new dominant pole is influenced by Generator 2 and all future improvements must be made to this generator.

Comparative Study of Single-Input and Dual-Input PSS

309

5 Conclusion Power system stabilizers are an integral part of the excitation circuit of the synchronous generator. In this paper, two types of PSSs are compared: PSS1A and PSS2A. The tuning of the PSS was performed using the pole placement method, while the location of the PSS was determined using the participation analysis, which revealed that the dominant oscillatory pole of the system is mainly influenced by Generator 1. After integrating the PSS, the dominant pole was damped, i.e., shifted to the left side of the S-plane, resulting in a more stable system. In addition to improving small-signal stability, the PSS1A implementation improved transient stability since oscillations observed in the time-domain were damped in a much shorter time after disturbance. Even better system response is achieved by implementing PSS2A where observed variables in the timedomain-generator active power, rotor angle and speed reach a steady state more quickly. The participation analysis after the PSS implementation showed the maximum influence of another synchronous generator on the new dominant oscillatory mode, hence further improvements must be made to this generator. PSS tuning is complex task and future work will be focused on applying machine learning methods in tuning process. Acknowledgment. This work was supported by Croatian Science Foundation under the project “Prosumer-rich distribution power network” (project number: UIP-2020-02-5796).

References 1. Sreedivya, K.M., Jeyanthy, P.A., Devaraj, D.: An effective AVR-PSS design for electromechanical oscillations damping in power system. In: 2019 IEEE International Conference on Clean Energy and Energy Efficient Electronics Circuit for Sustainable Development (INCCES), pp. 1–5. IEEE (2019). https://doi.org/10.1109/INCCES47820.2019.9167703 2. Du, P., Fang, X., Wang, Y., Yang, Y.: Sensitive load identification method based on voltage sag monitoring data. In: 2020 IEEE 4th Conference on Energy Internet and Energy System Integration (EI2), pp. 3940–3944. IEEE (2020). https://doi.org/10.1109/EI250167.2020.934 7294 3. Abou-El-Soud, A., Elbanna, S.H.A., Sabry, W.: A strong action power system stabilizer application in a multi-machine power system containing a nuclear power plant. In: 2019 16th Conference on Electrical Machines, Drives and Power Systems (ELMA), pp. 1–6. IEEE (2019). https://doi.org/10.1109/ELMA.2019.8771641 4. Li, Z.E., Tiong, T.C., Wong, K.I.: Improving transient stability of diesel-wind-solar hybrid power system by using PSS. In: 2019 1st International Conference on Electrical, Control and Instrumentation Engineering (ICECIE), pp. 1–6. IEEE (2019). https://doi.org/10.1109/ICE CIE47765.2019.8974702 5. Khalil, Z.E., El-Said Eliwa, A.E.-F., Sabry, W.: A design of a modified power system stabilizer for power system transient stability enhancement. In: 2018 Twentieth International Middle East Power Systems Conference (MEPCON), pp. 712–717. IEEE (2018). https://doi.org/10. 1109/MEPCON.2018.8635297 6. Alsakati, A.A., Vaithilingam, C.A., Alnasseir, J., Jagadeeshwaran, A.: Transient stability improvement of power system using power system stabilizer integrated with excitation system. In: 2021 11th IEEE International Conference on Control System, Computing and Engineering (ICCSCE), pp. 34–39. IEEE (2021). https://doi.org/10.1109/ICCSCE52189.2021.9530970

310

T. Košorog et al.

7. Liu, Z., et al.: Research the influence of PSS on power system transient stability. In: 2022 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA), pp. 134–138. IEEE (2022). https://doi.org/10.1109/ICAICA54878.2022.9844499 8. Patel, K.: Transient stability analysis and tuning of power system stabilizer for three machine nine bus system using frequency response approach. In: 2020 International Conference on Advances in Computing and Communication Engineering (ICACCE), pp. 1–6. IEEE (2020). https://doi.org/10.1109/ICACCE49060.2020.9155046 9. Nikolaev, N.: Tuning of power system stabilizer PSS3B and analysis of its properties. In: 2018 20th International Symposium on Electrical Apparatus and Technologies (SIELA), pp. 1–5. IEEE (2018). https://doi.org/10.1109/SIELA.2018.8447166 10. Sorrentino, E., Leon, F.: Comparison among typical input signals of different types of Power System Stabilizers (PSS). In: 2020 IEEE ANDESCON, pp. 1–6. IEEE (2020). https://doi. org/10.1109/ANDESCON50619.2020.9272090 11. Pourbeik, P., Vowles, D.J., Gibbard, M.J.: Small-Signal Stability, Control and Dynamic Performance of Power Systems; University of Adelaide Press, Adelaide, Australia (2015) 12. Kundur, P.: Power System Stability and Control. McGraw-Hilll, New York (1993) 13. IEEE Recommended Practice for Excitation System Models for Power System Stability Studies in IEEE Std 421.5-2005 (Revision of IEEE Std 421.5-1992) (2006). https://doi.org/10. 1109/IEEESTD.2006.99499 14. Larsen, E.V., Swann, D.A.: Applying power system stabilizers. Part I: General concepts. IEEE Power Eng. Rev. PER-1(6), 62–63 (1981). https://doi.org/10.1109/MPER.1981.5511615 15. Larsen, E.V., Swann, D.A.: Applying power system stabilizers. Part II: Performance objectives and tuning concepts. IEEE Trans. Power Apparatus Syst. PAS-100(6), 3025–3033 (1981). https://doi.org/10.1109/TPAS.1981.316410 16. Larsen, E.V., Swann, D.A.: Applying power system stabilizers. Part III: Practical considerations. IEEE Trans. Power Apparatus Syst. PAS-100(6), 3034–3046 (1981). https://doi.org/ 10.1109/TPAS.1981.316411 17. Kundur, P., et al.: Definition and classification of power system stability IEEE/CIGRE joint task force on stability terms and definitions. IEEE Trans. Power Syst. 19, 1387–1401 (2004)

Creation and Maintenance of Public Real Estate Records - Business Models in Croatia Milan Ivanovi´c1(B)

, Franjo Ambroš2 , and Vedran Stojnovi´c2

1 PANON Think Tank, Vijenac Ivana Meštrovi´ca 19, 31000 Osijek, Croatia

[email protected] 2 GEOprem Ltd., Trg Lava Mirskog 1, 31000 Osijek, Croatia

Abstract. The paper briefly describes the Croatian real estate registration system, its subjects and basic processes, and points to the insufficient correlation of official real estate records with current state, which for years has been a major obstacle to the growth of business, the realization of investments, as well the management of states and other public assets. The unsettled state of public real estate records creates legal uncertainty for citizens, private investors, public and state institutions. This problem is also expressed in other countries of Southeast Europe. Defining real estate is a complex technical task that results in the materialization of boundaries between cadastral parcels as the basic record of the real estate and their recording in the selected coordinate system. This work in Croatia is entrusted to private geodetic companies whose activity is regulated by law through public tenders. Preparation of tenders on behalf of the Government of the Republic of Croatia (for works related to public real estate) is carried out by the State Geodetic Administration, and for private investors and entrepreneurs, works are regulated under market conditions. The paper investigates business model for more efficient creation and maintenance of public real estate registers - with special reference to the economic position of geodetic activity, and in the conclusion proposes measures to overcome this situation. Keywords: Real estate · Real estate registers · Legal security · Salaries in geodesy · Business model

1 Introduction Creating and maintaining public real estate records appeared as a problem immediately after the democratic changes and the beginning of the process of post-socialist transition in the Republic of Croatia - at the beginning of 1990. It is about an insufficiently built system of public records, that is, about the economic problems of implementing these tasks. It was hypothesized that the constellations between the subjects of the real estate registration system in the Republic of Croatia are very unusual and illogical, i.e., they are not consistent either from the perspective of the market or from the point of view of the state administration. The paper: (a) analyzes the Croatian system of real estate records, (b) describes the Reform of land registers and cadastral surveys, (c) considers the state of real estate records in Croatia and the structure of geodetic activities in previous years © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 311–329, 2024. https://doi.org/10.1007/978-3-031-51494-4_26

312

M. Ivanovi´c et al.

and the current state. In particular, (d) the economic position of geodetic activity and tenders for geodetic work, (e) the market framework of geodetic work in the Republic of Croatia and (f) the economic position of employees of geodetic companies are analyzed. Likewise - the (g) business model of the geodetic company was analyzed in terms of operating costs, and conclusions were proposed.

2 Croatian Real Estate Registry System Public records on real estate Croatian are heritage from the Austro-Hungarian monarchy. By passing the “Imperial Patent” (Act on Cadastral Survey) on 23 December 1817, Emperor Franjo I. prescribed the procedures for real estate registration. This Act had only 26 articles, but the accompanying regulation passed in 1820 (Cadaster and Surveying Instruction) had 477 articles. This law was based on modern knowledge of real estate registration and was a continuation of the advanced cadaster and the promotion of human freedoms taken from the French Civil Code (Code civil), which was adopted on March 21, 1804 and served as the most advanced code of its time. The Civil Code is also the beginning of the standardization of civil liberties, equality of citizens before the law, freedom of religion, the right to civil marriage instead of church marriage. The basis of that law is the search for a “measure” between customary law in the north and written (Roman) law in the south of Europe [1]. Austria-Hungary started land surveys for the purpose of recording real estate on the territory of Croatia in Istria in 1818, followed by Dalmatia in 1823. The Hungarian part of the monarchy started with surveys after the intervention of the “central government” in 1849. The exception is Slavonia, which started the land survey in 1847 [2]. The continuity of measurements took place until 1877, when most of the monarchy was measured. In parallel, cadaster offices were formed with the task of recording changes to real estates. This survey had the purpose of introducing a fair tax on real estate, so a credit assessment of the land was also carried out. Since that time, the center of gravity of state financing has shifted from wars to tax collection. The data from that land measurement are still in official use today in about 70% of the territory of the Republic of Croatia [2]. Real estate ownership records in the Republic of Croatia are based on two public registers: (a) the Land Register - which is maintained by the municipal courts and (b) the Cadaster - which is managed by the State Geodetic Administration (SGA). - The Cadaster is an official record that contains spatial data on parcels - cadastral parcels (position, shape, area), data on land use as well as data on users (land possessioners). The SGA takes care of cadastral data through 20 regional cadaster offices with 92 branches, and the City Office for Cadaster and Geodetic Affairs of the City of Zagreb. - The Land register is a record of the legal status of real estate. Real and legal rights to real estate are stored in the land registers. Land registers should be the basis of legal certainty in real estate transactions. Land registers are under the jurisdiction of the Ministry of Justice and Administration, and 108 land register departments at the Municipal Courts take care of them.

Creation and Maintenance of Public Real Estate Records

313

2.1 Land Registry Reform Since 2003, the Government of the Republic of Croatia has been implementing the National Program for the Improvement of the Organization of Land Registers and Cadasters (the project “Uredena zemlja”) through the Ministry of Justice and Administration and the SGA. This program includes all activities undertaken by the Ministry responsible for judicial affairs and the SGA to modernize and improve the way real estate is registered in Croatia. In addition to regular activities and a number of special projects, one of the key components of the reform is the project to improve the organization of land registers and cadasters. The project was launched with the goal of establishing an effective land administration system that will contribute to the development of a wellfunctioning real estate market. Within this framework, the Joint Information System of Land Registers and Cadasters (ZIS) was developed, the goal of which is to establish a unique database and applications for managing and maintaining cadaster and land register data. ZIS has been in use since November 21, 2016 at all 108 land registry offices and 113 cadaster offices. The modernization of the cadaster and land registers in Croatia accelerated and simplified the process of registering real estate and related ownership rights. Data can be obtained from the cadaster and land records immediately; all cadastral and land registry data are digitized and available on the Internet 24 h a day. This system enables the electronic issuance of extracts from land registers (via the e-Citizens platform) and the electronic submission of requests to authorized users (lawyers, notaries public and state attorneys) for registration in the land register [3]. 2.2 Cadastral Surveys The cadastral survey collects and processes all necessary data for the establishment of cadastral parcels, registration of buildings, recording of special legal regimes on land and ways of using land, and preparation of cadastral records. The cadastral survey for the cadastral municipality is carried out by the SGA in agreement with the ministry responsible for judicial affairs, and certain tasks within the framework of the cadastral survey are performed by authorized private geodetic companies through public tenders. According to article 8 of the Law on State Survey and Real Estate Cadaster [4], jobs on state survey and real estate cadaster are performed on the basis of multi-year and annual programs. The program determines the areas where cadastral surveys will be carried out, as well as sources of funding for the implementation of the program. The funds needed to execute the program are provided in the state budget or from other sources. Local self-government units can include in the financing of real estate cadaster operations legal entities and natural persons who are holders of rights to real estate in the area of performance of these operations. When all the necessary data are collected and processed by the cadastral survey, the cadastral survey elaboration report is prepared. In addition to other parts, the cadastral survey report must contain a cadastral plan and title deed on every parcel. On the cadastral plan, the cadastral parcels are shown in such a way that their borders, the buildings built on them and the numbers of the cadastral parcels are visible. The cadastral plan also shows the house numbers and the boundaries of different ways of use on the cadastral parcel. In the title deeds, all collected and processed data on a cadastral parcel is shown - data on holders of real estate rights collected on the basis

314

M. Ivanovi´c et al.

of available documents (land register and cadaster), as well as statements of interested parties are reported [5].

3 Situation in the Real Estate Registry in the Republic of Croatia A state as a subject of international law should have the following characteristics: • • • •

Permanent population Defined territory Government Ability to create relations with other countries

The territory of the state consists of real estate owned by it, owned by legal entities of which it is the founder, owned by local and regional self-government units, and owned by legal entities and natural persons. The basic record unit of the land cadaster in Croatia is the cadastral parcel. The owners of these cadastral parcels are not necessarily permanent residents of the state, so relations with other states must be arranged in order to use the real estate efficiently. A disorganized data in system of real estate records is a potential generator of problems for all owners. Consequently, it generates legal uncertainty for the entire country. As mentioned in the introduction, the cadastral survey was initiated by the state, and this principle is still valid today. Owners are not considered competent to initiate geodetic surveys that would result in regulated real estate. The former state directed cadastral surveys to areas that were not part of the Austro-Hungarian Monarchy because they did not have real estate records. The area of eastern Croatia was regulated through the process of consolidation of agricultural land, while the rest of Croatia relies on the Austro-Hungarian survey. In the past 200 years, the owners sporadically participated in the surveys. The state did not think strategically about this problem and provided a budget for continuous measurement. Today, unkempt real estate on the one hand repels investors, and on the other hand, favors real estate half legal brokerage. Croatia has around 14.4 million cadastral parcels; we estimate that around 60 million holders are registered on them. Ownership is a constitutional category in Croatia and as such has the highest legal protection in the country. Restrictions arise from the need to use real estate for the public good, so in special cases rights can be limited. These restrictions are regulated by law. For each ownership restriction, the owner is entitled to compensation according to the nature of the restriction or the extent of the expropriation. 3.1 Insufficient Synchronization of Real Estate Records In a series of published professional papers and texts in recent years, several authors have pointed out the problems of insufficient synchronization of real estate records between Land register and Cadaster, and synchronization between official records and real situation in the field. We selected some of them: (a) “Seven out of 10 investments are on hold due to chaos in the land records” - points out (2017) Suzana Varošanec: “This problem is also reflected in the real estate market: out of ten transactions that are negotiated, only one in three is finalized, agents warn.

Creation and Maintenance of Public Real Estate Records

315

Unsettled land records cause significant damage to the state and local community, potential buyers and sellers, and make it difficult for real estate brokerage agencies to work, and are one of the causes of the slow functioning of the real estate market. According to the data of the study prepared for the purposes of the Project to introduce the Cadaster of Buildings in the Republic of Croatia, there are 3.1 million buildings with 1.5 million apartments worth around 10 billion euros. However, due to unsettled records on buildings and apartments, it is estimated that the income from communal fees is lower by about 20%, which is a very significant loss of money. In the same way, this problem is reflected in the real estate market: out of ten transactions that are negotiated, every third succeeds, and in other cases, real estate agencies face either the problem of mismatches in the land registers and the cadaster, or with properties that are not condominiums [6]. (b) Vice President of the Government of the Republic of Croatia - Anja Šimpraga also indicates (2021) that “The land register and the cadaster are not harmonized; in some parts of Croatia, the land registers are not harmonized, so the areas do not correspond to the actual state of affairs. The owners are different, some have died, the areas are not unified, the property descriptions are outdated [7]. (c) Vedran Vukobrat (2022) asks “Why do local units still do not know what assets they have?” [8] We believe that the findings that public real estate records are not in order are based on the following facts: 1. The boundaries of cadastral parcels in public records do not match the boundaries on the ground, 2. Special legal regimes on cadastral parcels are not spatially demarcated, 3. The address of the cadastral parcel is not current, 4. Land use culture was not recorded (actual use of the cadastral parcel), 5. The registration of buildings and other facilities on the cadastral parcel is incomplete, 6. There is no information on legality for buildings, 7. A building plot has not been formed in accordance with the spatial plan, 8. For the owners/occupiers, information on the identifier (OIB) and address of residence is missing, 9. In reality, secret divisions exist, while in public records the owners are recorded as co-owners, 10. For a large number of buildings, the floor plan procedure was not carried out or the Land registers main book was not connected with the book of submitted contracts, 11. Building projections are not recorded on cadastral plan if their height from the ground is greater than 4 m, 12. Registration in the Land register is not provided for all ownership restrictions (i.e., “right of way” for electronic communication infrastructure). 3.2 How to Arrange Real Estate Real estate arrangement is imperative for the further development of the country. For a long time, there was a belief in Croatia that real estate records could be gradually improved. Geodetic surveys were considered to be very complex operations, so they

316

M. Ivanovi´c et al.

were carried out sporadically. In addition, the renewal of the land register that followed was not organized at the same time as the measurements, so the procedure took a long time. Improvements in records were reflected in the procedures of “matching” and “homogenization of cadastral plans”. The result of these principles is the change of real estate boundaries during every technical intervention that required a geodetic survey of the actual situation. Until 2010, Croatia had two projected coordinate systems based on the Gauss-Krüger projection, and it was based on local geodetic datum called “HDKS” which used a Bessel ellipsoid. The basis for the measurement was permanent geodetic points (trigonometers, polygonal points…). The final results of the measurements were the result of the adjustment of the entire area. It was only in 2010 that the Decision on the introduction of a new unified projection of the Republic of Croatia (HTRS96/TM projection for position (based on global geodetic datum and synced with ETRS89 datum) and HVRS71 for heights) entered into force [9]. Use of global datum made easier adoption to new GNSS (Global Navigation Satellite Systems) technology. With the introduction of the CROPOS service in 2008 (Croatian Positioning System), geodetic measurements could be performed without the need for data adjustment. Originally, the measurement was made using two GNSS receivers, connected by a radio link to correct data received from GNSS satellites. By building permanent GNSS base stations and sending corrective parameters via Internet protocol to GNSS devices in motion (establishment of CORS network), the efficiency of field measurements has been significantly increased. The paper [10] investigated the problem of Internet service availability in rural areas and suggested recommendations on how to invest in this infrastructure in order to increase the productivity of geodetic surveying. The Law on Land Registry [11] prescribes the method of individual land registry correct procedure, so with the previous technical and institutional conditions, it is possible to harmonize the situation on the ground with data in public records. The only condition is that the neighbors agree with the position of the intermediate points, which they declare in a special administrative procedure. The third element of the regulated boundary in accordance with the Law on Property and Other Real Rights should also be included in the law. The coordinates of property boundary should have the same weight (or even bigger) than the physical boundary mark (boundary stone, fence, etc.). By destroying the boundary marker, the relevant coordinate is stored in the Cadaster. In this way, the procedures for border regulation would be reduced to the renewal of the border mark according to the cadaster records, which the geodetic contractor would perform at the request of any party [12].

4 Geodetic Activities Geodesy is a science that deals with measuring and displaying the Earth’s surface, determining the Earth’s shape and its gravity field. The predominant subject of activity in geodesy is the geodetic survey of land for the needs of the real estate cadaster. The basic branches of geodesy are satellite, physical, maritime and hydrographic geodesy, engineering geodesy, photogrammetry and remote sensing, cartography and geoinformatics. Geodesy is widely used in construction, mining, agriculture, forestry, environmental protection, protection of cultural monuments, spatial planning, urbanism, national security and defense, shipbuilding, industry, medicine and elsewhere [13].

Creation and Maintenance of Public Real Estate Records

317

4.1 Organization of Geodetic Activities Geodetic activities, i.e., procedures that result in geopositioning that affects public records, are regulated in Croatia by the Law on Performing Geodetic Activities [14]. Geodetic work in Croatia is performed by authorized geodetic engineers (VIIth degree of education), and they are assisted by professional associates (VII. /VIth degree) and associates of authorized geodetic engineers (Vth degree). On the basis of that law, the Croatian Chamber of Chartered Geodetic Engineers was founded with the purpose of controlling geodetic work as work of interest to the state. SGA has control over the work of the Croatian Chamber of Chartered Geodetic Engineers. Organizationally, SGA is under the jurisdiction of the Ministry of Spatial Planning, Construction and State Property. This complex structure also points to problems, which are reflected in the diverse actions of these organs. The fact is that the legal norms in procedures for maintaining data on real estate have often changed since the independence of the state, especially since its admission to the European Union, as a result of which geodetic contractors must change their procedures in accordance with the new regulations. All this requires constant education from geodetic contractors. This training is organized by the Croatian Chamber of Chartered Geodetic Engineers, and is conducted by institutions, associations, companies or individuals in accordance with the annual training plan. The choice of topics for education is left to each geodetic contractor. The task of SGA is the organization of state surveying (providing a reference system, setting and maintaining basic geodetic points, determining the parameters of the earth’s gravity, gravimetric and magnetometric measurements, topographic surveying and making state maps, as well as marking and keeping records of the state border), real estate cadaster (cadastral spatial units, cadastral measurements, preparation and maintenance of real estate cadaster operations), register of buildings, cadaster of infrastructure, preparation and maintenance of a joint information system of cadaster and land registers, management of the Register of Geographical Names, organization of the archive of geodetic studies, administrative tasks in the area of its competence and carries out geodetic inspection control. Jobs are performed in accordance with annual and multi-year plans. The State Geodetic Administration, as an umbrella institution, organizes, supervises and secures the budget for tasks within its scope. 4.2 State and Circumstances of Previous Years On the state of geodetic activity in Croatia at the beginning of the 21st century, writes Hori Martini´c: In October 2001, 440 companies, trades, CGE offices (CGE = Chartered geodetic engineer) and institutions received approval from the SGA to perform state surveying and real estate cadaster work. In order to better inform the entities in the system, the State Geodetic Administration published the publication “Geodetski informator” (Geodetic informant) with data on the authorized persons - which went through two more editions. In the same year, the SGA geodetic inspection began to work for the purpose of: resolving disputes, handling applications, and verifying the conditions and methods of performing state surveying and real estate cadaster work for individuals and legal entities. The number of CGE companies and offices, institutions and public companies authorized for state surveying and real estate cadaster work in 2004 is around

318

M. Ivanovi´c et al.

500 and is constantly growing. In recent years, the scope of work in geodesy has been increasing. The construction industry has recorded high growth rates in recent years, which is reflected in the increased demand for geodetic services in civil engineering. However, it is noted that the number of authorized companies is growing faster than the total number of employed surveyors. There is a tendency to decrease the number of experts in geodetic companies. There seems to be a strong philosophy that everyone wants to be their own boss. There have been cases of smaller branches joining together for the purpose of more efficient operations, which ended in disassociation. Smaller entities are more dynamic, respond better to market demands and more easily adapt to turbulent changes in the economy of a country in transition, such as Croatia. However, the market is also looking for large and medium-sized companies, so individual and ambitious owners with such visions, equipping and hiring new workers, began to realize their ideas. Today’s circumstances are in their hands, given that there is enough work. It has been shown that in the case of larger jobs of national importance, the tradition and references possessed by “old” and/or “big” geodetic companies provide a guarantee that the job will actually be completed within the given period; you can trust their product, because they “will not play with” their reputation. Also, within large companies there is internal control and the work of such a company passes a certain filter within the company itself. A branch in which only the owner works, puts out the result of his work in the form in which it was originally made [15]. Research at the level of the European Union (2007) indicates that the number of surveyors in the country depends on the size of the market and the amount of work it creates. For example, in a number of Eastern European countries, where restitution of property rights, land privatization and real estate cadaster are still in development, the number of active surveyors exceeds the number of cadastral surveyors in Western European countries. Due to the different legal frameworks that affect the national cadastral system, as well as the domestic real estate market, the situation in each country is unique [16]. Figure 1 shows the number of CGE by counties in the Republic of Croatia (2012); there is a large concentration of surveyors in the coastal counties and the city of Zagreb [17]. Resources of the private geodetic sector in European countries (according to 2012 – Fig. 2 and 3): Germany has 2,600 authorized companies per 82 million inhabitants, i.e., 32 authorized companies per million inhabitants. This factor is the smallest in Great Britain, where it amounts to 3 authorized companies/thousand inhabitants. The largest is in Slovakia: 335 companies/thousand inhabitants. According to this factor, Croatia is right behind Slovakia with the number of 131 companies/thousand inhabitants. Regarding the number of authorized engineers per million inhabitants, Germany has 1,482 authorized engineers and is also at the bottom of the ranking with a ratio of 18 CGE/mill. populations. Luxembourg has the lowest ratio (12). Croatia ranks third with 166 CGE /mill. inhabitants, behind Denmark (260) and Estonia (179). In the case of Croatia, this factor is the result of the expansion of both authorizations and the establishment of companies in the transition period due to the increased need and pressure from the EU for a quick and efficient settlement of land issues. However, recently, mostly due to economic situation and economic crisis, but also due to saturation of the market with the number of authorized companies and engineers, there is noticeable stagnation and a slow growth trend of the aforementioned factors [18].

Creation and Maintenance of Public Real Estate Records

319

Fig. 1. Number of active CGEs - March 27, 2012 in Croatia - by county [17].

Fig. 2. Number of authorized companies and CGE per population per country (2012) [18]

Fig. 3. Number of authorized companies and CGEs per mill. inhabitants of Southeast Europe (2012) [18]

In Croatia, the number of CGEs per population is 1.6 per thousand, with this factor being the highest in the region (Fig. 3). The number of private companies per inhabitant

320

M. Ivanovi´c et al.

is also the highest in Croatia, with only Macedonia and Serbia showing a similar trend: small companies predominate, and the ratio of the number of authorized offices to the number of authorized persons is almost one to one, while in the rest of the region this factor is below 1. In the weakest geodetic regulated country, Bosnia and Herzegovina, only the beginning of the development of land administration in the hands of the private sector is visible, and the licensing system is only in its theoretical infancy. In the Republic of Croatia, the issue of the number of CGEs is raised, which is too large (171 per million inhabitants), as well as the territorial distribution of CGEs, 25% of which are concentrated in Zagreb. Also, a big problem already is the surplus of geodetic technicians [18].

Fig. 4. Analysis of the number of cadastral companies by revenue in Croatia by county in 2016 [19]

The total market value of the geodetic profession in 2016, i.e., the amount of income of geodetic companies registered with HKOIG, according to the calculation in this paper, was slightly less than HRK 525 million. The amount of the company’s total income varies significantly depending on the location and area of activity of the company. The city of Zagreb, Split and Rijeka counties together share more than 50% of the market, while companies from the ten counties with the lowest revenues share about 10% of the market. The largest revenues from geodetic activity are realized in the City of Zagreb and in large regional centers (e.g., in Osijek-Baranja County) and in coastal counties with the exception of Lika-Senj County, Fig. 4 [19]. 4.3 Current Situation (2023) The current state of CGEs in the Republic of Croatia (2023) - according to the counties is shown in Table 1 and Figs. 5 and 6. The city of Zagreb and the counties of Split and Rijeka and the coastal counties have the largest number of CGEs and associates, and the largest number are geodetic companies (54 in total) with two employees.

Creation and Maintenance of Public Real Estate Records

321

Fig. 5. Number of authorized cadastral companies by number of employed surveyors in Croatia [20] Table 1. Employed surveyors by types of legal entities - by counties [22] No.

County

CGE/ companies

CGE/ office

CGE/ common office

Nr. profess. associates

1.

Bjelovarska

17

2

0

9

2.

Brodska

14

3

0

17

3.

Dubrovaˇcka

32

6

0

7

4.

Grad Zagreb

333

5

0

148

5.

Istarska

76

15

0

47

6.

Karlovaˇcka

34

4

0

11

7.

Koprivniˇcka

10

3

0

6

8.

Krapinska

18

4

0

10

9.

Liˇcka

8

3

0

4

10.

Medimurska

25

0

0

12

11.

Osjeˇcka

41

9

0

58

12.

Požeška

11

2

0

6

13.

Primorska

94

5

0

81

14.

Sisaˇcka

15

3

2

11

15.

Splitska

148

13

0

79

16.

Šibenska

29

11

0

14

17.

Varaždinska

22

5

0

5 (continued)

322

M. Ivanovi´c et al. Table 1. (continued)

No.

County

CGE/ companies

CGE/ office

CGE/ common office

Nr. profess. associates

18.

Virovitiˇcka

6

2

0

2

19.

Vukovarska

15

4

0

11

20.

Zadarska

53

15

2

18

21.

Zagrebaˇcka

44

6

0

40

Total

1045

120

4

596

Fig. 6. Number of CGE employees and professional associates in Croatia 2023 - by county. [22]

5 Economic Position of Geodetic Activity Geodetic survey work in Croatia is carried out by private geodetic companies whose activities are regulated by law. Cadastral surveys are contracted through a public tender conducted by the State Geodetic Administration, financed from the state budget, and for private investors and entrepreneurs, jobs are contracted under market conditions. According to the experiential assessments of experts in geodetic affairs in Croatia. Small surveying companies perform about 80% of their work for natural persons and 20% for legal entities. Large geodetic companies do the opposite - 80% of their business is with public institutions and legal entities, and 20% with natural persons. In the case of business with natural persons, business relations are determined by the local market, and in the case of business with legal entities, a tender or negotiation on the participation of the volume of business in the total investment. Large companies that have a need for everyday geodetic services form their own geodetic teams that appear on the geodetic market from time to time.

Creation and Maintenance of Public Real Estate Records

323

5.1 Tenders for Land Surveying Jobs Financing of geodetic surveys is done with funds from the budget for geodetic activities. The impossibility of continuously securing the budget for geodetic surveys was the reason that the cadastral surveys were not carried out continuously. In addition, the budget for the renewal of the land register was planned by the Ministry of Justice and Administration, and the funds were provided by the Ministry of Finance. The lack of coordination in budget planning resulted in delays in real estate development. At the same time, the document [19] states that the process of dealing with legal rights and creating new Land register records for the owners (exposure of geodetic elaborate took 1–10 years) is the main cause of the long deadline for finishing entire job (each tender for the entire cadastral municipality) and putting new records into active state. All real estate development activities planned by the state were contracted in accordance with public procurement procedures [21]. Access to that service market was aimed at larger geodetic companies, which were also going through the process of transition, changes in the ownership structure or changes in generations of experts. The assessment is that in the tenders, the cost estimates were deficient for the prescribed new administrative procedure (lack of experience on both sides), so the subsequent work significantly reduced the expected profit from the contracted work (Fig. 7).

Fig. 7. Budget for geodetic surveys in Croatia 2004–2011 (in mill. e) [22].

Public procurement procedures and a reduced market for geodetic services resulted in constant price reduction at the expense of quality, investment in equipment and reduced training of professional staff. 5.2 Market Framework of Geodetic Works in the Republic of Croatia The constellations of relations between the subjects of the real estate registration system in the Republic of Croatia are very unusual and illogical. On the one hand, for three decades (since the beginning of the process of post-socialist transition from a planned economy and a centralized state to a multi-party, civil society, market economy and a legal state with the separation of powers) there have been pronounced problems with

324

M. Ivanovi´c et al.

the arrangement of public real estate records (Land Registers and Cadaster), that is there is a great need for geodetic measurements (a kind of market demand), and on the other hand, there are expert teams (numerous private geodetic companies) that can solve mentioned problems, but - the situation is not being solved satisfactorily. It should be emphasized here that - in addition to the need to solve the backlog from the socialist period - modern processes in society, the economy and the state constantly produce new needs for geodetic services. In Croatia, there is an oligopoly of public infrastructure companies (Hrvatska elektroprivreda, Hrvatske vode, Hrvatske šume, Hrvatske ceste, Hrvatske željeznice) and the State Geodetic Administration - which, by insufficient allocation of funds for the needs of geodetic surveys and at the same time by dictating low, inappropriate prices for geodetic surveying jobs, are blackmailing numerous disunited private geodetic companies. The surveying companies themselves contributed to this situation, because they often accepted jobs below the realistic prices of the labor used. Thus, they contributed to the reduction of the prices of geodetic services, which was not accompanied by productivity, and all this was reflected in the quality of services. For years, these processes adversely affect the dynamics of arranging public records on real estate - with significant financial losses to the local self-government and the state budget, and slow down development processes and contribute to an unfavorable political and social climate in the country and abroad. At the same time, these processes put geodetic companies in an increasingly unfavorable economic position and slow down the technical progress and equipping of these companies with modern geodetic and IT equipment and IT support. The aforementioned oligopoly constellations have already resulted in a series of disadvantages and the situation will become more unfavorable every day. 5.3 Economic Position of Employees of Surveying Companies There are few works in the Republic of Croatia on economic processes and the economic position of geodetic activity. The results of the two analyzes will be briefly stated here. The analysis of the income of geodetic companies registered with the Croatian Chamber of Chartered Geodetic Engineers (HKOIG) was carried out on the basis of data available on the online portal Fininfo, i.e., the presentation of the annual market activity of geodetic companies from 2012 to 2016. Data on the value of work in the profession, i.e., the average gross/net salaries of employees, are also available as public data within the Fina (Financial Agency) service. However, the average salary in a company does not allow for more detailed analysis according to the employee’s work experience, i.e., level of education or other criteria, as well as benefits. In order to enable such an analysis, data available through the “Moja Pla´ca” (My salary) service were used, which were obtained by surveying employees employed in the profession according to their individual characteristics - Fig. 8 [19]. The authors of this paper also conducted internal research on salaries in geodetic companies in Croatia in the period 2013–2017 - based on data from the online portal Bisnode - Fig. 9. So, in that period, the average salary of employees in geodetic companies in Croatia was lower than the average salary for all employees in the Republic of Croatia. – except for 2016. It should be remembered here that surveying companies

Creation and Maintenance of Public Real Estate Records

325

Fig. 8. Average net salaries of employees in geodetic companies with regard to work experience and level of education (data provided from the My salary portal in e) [19]

Fig. 9. Net salary of employees in geodesy companies compared to the average salary in Croatia in e [20].

have a more favorable qualification structure and perform more complex tasks than the average employees in Croatia) [20].

6 Business Model of the Geodetic Company As stated before - geodetic activity in Croatia is a profession regulated by law. Professionals go through procedure of training, work under supervision, taking exams and constant annual renewal or supplementing of knowledge. Renewal of knowledge is multidisciplinary. The Croatian Chamber of Chartered Geodetic Engineers provides a wide range of topics. The education is decentralized and is intended as an exchange of experiences for the purpose of business association in case of need. In addition to experts,

326

M. Ivanovi´c et al.

the surveying company employs administrative staff and, if necessary, support staff. The equipment with which company carries out its work is diverse: geodetic equipment for various measurement purposes, IT equipment with high processing power for processing large amounts of data, software support for processing various digital data, including processing of image records, vehicle fleet for transporting people, including transport on rivers, lakes and the sea. For the use of licensed rights (software), company pays a one-time fee or periodical fee upon agreed amount. For a multi-day stay in the field, the company bears the costs of accommodation and per diem, for larger work sites, office space must also be rented on the field. For employees, the company must provide adequately equipped business premises with broadband access (optical Internet). It is estimated that a net space of approx. 12 m2 should be provided per employee. 6.1 Business Expenses of the Geodetic Company In designing the cost model of an average surveying company, the starting point is the continuous utilization of working time of 80% (annual pool of 1920 h and the following groups of costs): 1. The cost of a professional staff position • • • • • •

depreciation of office equipment - 4 years. car depreciation - 5 years. - the car is used by two experts instruments depreciation 5 years - equipment for 2 experts office space - rent 10 euros/m2 cost of education per employee per diem for work in the field - an average of 5 per month

2. The cost of an administrative staff position • • • • •

depreciation of office equipment - 4 years. office space - rent 10 euros/m2 cost of education per employee – 10% of the amount the cost of office materials the cost of office space for the archive

3. Gross salary of professional staff • net payment to the employee • compensation for transportation costs • food allowance 4. Gross salary of administrative staff • net payment to the employee • compensation for transportation costs • food allowance The estimate of the cost of a professional staff position (without administrative staff) reduced to a unit hour is 30 euros/hour. The estimate of the cost of the workplace increased by the cost of administrative staff (approx. 30%) is 39 euros/hour. This calculation does

Creation and Maintenance of Public Real Estate Records

327

not include taxes paid by the company or cost of capital (interest expense and profit payout). A small company (1–2 employees), which has no administrative staff and is not in the value added tax system, offers services on the market at a price of 24 euros/hour. 6.2 Employee Motivation Continuous activity is looking for a motivated employee. Salary is certainly one of the main elements of a satisfied employee, but it is not the only one. An orderly work environment, relations with superiors, accommodating colleagues, the possibility of advancement and self-affirmation can be a tip on the scale of motivation. Unfortunately, all this requires taking care of fairness and insisting on the realization of planned business goals. 6.3 How to Run a Business in Unfavorable Conditions? For several years now, managers of geodetic companies have had a basic task - to survive this unfavorable situation, preserve professional staff and maintain their position on the market. There is little space left for professional training, development of new technologies, new services and expansion into new markets. In such a situation, jobs are accepted where there is a financial loss, i.e., to the detriment of the economic position of the company and its employees.

7 Concluding Remarks • Creation and maintenance of public real estate records in the Republic of Croatia is based on two public registers: (a) Land Register - managed by municipal courts and (b) Cadaster - managed by the State Geodetic Administration (SGA). • The geodetic services market, in addition to the creation and maintenance of the real estate register, also includes other registers: infrastructure cadaster, polluter cadaster, building records, state border records. The segment of engineering geodesy is closely related to the design and construction of buildings, construction supervision, control of exploitation of mineral raw materials, documentation of immovable cultural heritage and other segments of activity. • The needs of the state and citizens require georeferenced data, as a supplement to existing data. The principle of free access to state data increases insight into the mismatch between the state of the field and public records, so the issue of state registers for the legal security of both the state and its citizens is raised. • Geodetic companies are carriers of geodetic measurements for the purpose of creating public real estate registers. A diagnosis of their operations is necessary in order to establish and maintain the partnership relationship of all participants, the state, local and regional communities, legal and natural persons, educational institutions, the Chamber of Chartered Geodetic Engineers and the companies themselves. • Through an active dialogue of all stakeholders, a new trust should be built that was broken by the oligopoly, the result of which is the inadequate price of geodetic services and the disunity of geodetic companies.

328

M. Ivanovi´c et al.

• Under such conditions, it is impossible to set up development business models for economic entities as well as for state institutions and the population. A broad professional, social and political discussion needs to be started on these problems. • In Croatia, there were no real estate records management projects financed from European Union funds, with the explanation that real estate management belongs to the private sphere. In this way, the main intention of the Republic of Croatia, the entry into the European Union and the application of the rule of law, is delayed, because unorganized real estate generates legal uncertainty for the state and the owners of real estate on the territory of the state.

References 1. Cro Encyclopedia. https://www.enciklopedija.hr/natuknica.aspx?id=12170. Accessed 7 July 2023 2. Roi´c, M., Paar, R.: 200 godina katastra u Hrvatskoj, VI hrvatski kongres o katastru, Hrvatsko geodetsko društvo, Zagreb, pp. 37–50 (2018) 3. Croatia’s system for registering real estate. https://e-justice.europa.eu/109/EN/land_regi sters_in_eu_countries. Accessed 7 July 2023 4. Zakon o državnoj izmjeri i katastru nekretnina, NN 112/18 5. Homepage of “Uredena zemlja” (Organized land) project. https://oss.uredjenazemlja.hr/. Accessed 7 July 2023 6. Varošanec, S.: Sedam od 10 investicija na ledu zbog kaosa u zemljišnoj evidenciji; Poslovni.hr (2017). https://www.poslovni.hr/nekretnine/sedam-od-10-investicija-na-ledu-zbog-kaosa-uzemljisnoj-evidenciji-326014. Accessed 7 July 2023 - (2021). https://www.portalnovosti.com/gru 7. Kožul, A.: Gruntovnica i katastar nisu uskladeni ntovnica-i-katastar-nisu-uskladeni. Accessed 7 July 2023 8. Vukobrat, V.: Zašto lokalne jedinice i dalje ne znaju kojom imovinom raspolažu? Informator 6751 (2022). https://informator.hr/strucni-clanci/zasto-lokalne-jedinice-i-dalje-ne-znajukojom-imovinom-raspolazu. Accessed 7 July 2023 9. Odluka o utvrdivanju službenih geodetskih datuma i ravninskih kartografskih projekcija Republike Hrvatske, NN 110/04 10. Toplek, G., Bilajbegovi´c, D., Štimac, M., Ambroš, F.: Mobilni Internet u funkciji CROPOSA - a i pove´canju geodetske produktivnosti, 3. CROPOS konferencija, DGU Zagreb, Opatija (2013) 11. Zakon o zemljišnim knjigama, NN 63/19 12. Ambroš, F., Aralica, T., Bajt, J., Gjurani´c, M., Konˇci´c, A.M.: Može li princip zaštite povjerenja u zemljišnu knjigu opstati bez zaštite povjerenja u katastar, 10. simpozij ovlaštenih inženjera geodezije, Hrvatska komora ovlaštenih inženjera geodezije, Opatija (2017) 13. Lapaine, M.: Geodezija u Hrvatskoj, Godišnjak Akademije tehniˇckih znanosti Hrvatske (2019) 14. Zakon o obavljanju geodetske djelatnosti, NN 25/18 15. Martini´c, H.: Obavljanje geodetskih poslova; Vijesti Državne geodetske uprave, Geodetski list br. 3/2004 (2004) 16. European requirements for cadastral surveyor activities. https://www.clge.eu/wp-content/upl oads/2008/04/european_requirements_for_cadastral_surveyor_activity.pdf. Accessed 7 July 2023 17. Hrvatska geodezija u brojkama, Ekscentar, br. 15, pp. 135–141 (2012)

Creation and Maintenance of Public Real Estate Records

329

18. Bjelotomi´c, O., Markovinovi´c, D., Klekovi´c, B., Baši´c, T.: Hrvatska geodezija na pragu EU; Zbornik HKOIG, V. Simpozij ovlaštenih inženjera geodezije, Opatija (2012). ISBN 978-953-55915-2-8 19. Baši´c, T., Masteli´c-Ivi´c, S., Grgi´c, M., Varga, M.: Geodezija i ekonomija - koliko je tržište geodetske struke u Hrvatskoj? Zbornik radova, Geodezija i drugi, 11. simpozij ovlaštenih inženjera geodezije, Opatija, pp. 181–186 (2018) 20. GEOPREM Osijek - prema evidencijama Bisnode 21. Zakon o javnoj nabavi NN 120/16, 114/22 22. HKOIG Homepage, Poboljšanje modela katastarskih izmjera (2014). https://hkoig.hr/images/ Pdf/Objasnjenja/T4-Dokument-HKOIG-katastar-Finalno-180314.pdf. Accessed 7 July 2023

Preventive Maintenance of Hydraulic Press Using Magnetic Testing Ljiljana Radovanovi´c(B) , Borivoj Novakovi´c , Mi´ca Djurdjev , and Luka Djordjevi´c Technical Faculty “Mihajlo Pupin”, University of Novi Sad, Zrenjanin, Serbia [email protected]

Abstract. This paper presents the application of magnetic testing as one of the non-destructive methods within the preventive maintenance activities of technical systems. Magnetic testing is performed on a hydraulic press. This research aims to timely detect potential defects in the technical system, the hydraulic press, to prevent large-scale failures. Keywords: preventive maintenance · nondestructive methods · magnetic testing · hydraulic press

1 Introduction Non-destructive testing or defectoscopy involves the development of technologies for detecting errors and assessing the impact of these errors on the material quality of system components. Non-destructive testing methods are based on the physical properties of the material being examined. Each method is designed to detect errors that share a common characteristic or perform specific measurements. These methods are suitable for predicting the reliability level of a system. The main advantages of non-destructive testing methods are the following: • they can be carried out directly on a part or a structure, regardless of its price, and without affecting its functionality, • it is possible to implement 100% control, • the sample is representative, • the same object can be tested using various methods, or the test can be repeated, • objects can be controlled during exploitation, • cumulative effects of errors can be monitored, • the mechanism of breaking of parts of the system can be followed, • preparation of the object for control is not required; except for the cleaning, • control can often be performed without stopping the working operations, • the testing equipment is portable (testing instruments).

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 330–341, 2024. https://doi.org/10.1007/978-3-031-51494-4_27

Preventive Maintenance of Hydraulic Press Using Magnetic Testing

331

Main drawbacks: • properties of the object (technical system) are measured indirectly in most cases, • certain methods require enhanced occupational safety, • interpretation of control results requires appropriately trained staff. Since non-destructive testing methods are based primarily on the physical properties of materials, the logical classification of methods according to the physical properties is proposed. This classification distinguishes ten main groups of methods (procedures): 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

sound methods, capillarity methods, magnetic methods, optical methods, radiation methods, radio wave methods, thermal methods, flow methods, electrical methods, electromagnetic methods [1].

Non-destructive testing (NDT), as a widely recognized term appearing in various publications and media, refers to a range of techniques encompassing NDI (nondestructive inspection) or NDE (non-destructive evaluation). The NDT methods ensure that surface quality, material properties, and geometry of tested components remain unaffected during testing processes [2]. Testing hydraulic systems can be complex and extensive for many reasons that can lead to potential failure. The test refers to system components, drive, control, and testing of the working fluid, i.e., oil. It is advisable to minimize or completely avoid water contamination in hydraulic systems [3]. Regarding hydraulic parameters, more complex ways and methods are presented in publications [4, 5]. Also, in the paper [6], Todi´c et al. examined the water-hydraulic parameters of the piston-axial pump.

2 Magnetic Tests Magnetic tests determine errors on the surface or below the object’s surface - methods for locating surface and internal discontinuities in ferromagnetic materials (pores, cracks, foreign materials, etc.). Only magnetic (ferromagnetic) materials such as iron, nickel, cobalt, carbon, low-alloy steels, and some alloy steels can be tested. Magnetization of the object can be performed before or during the test. Before the test, those objects are magnetized where the residual magnetism is sufficient to enable the test. During the test, objects with insufficient residual magnetism to perform the test and objects of large dimensions and complex shapes are magnetized. Assuming there is a crack in the object, a notch, or a sharp transition in the object, while a magnetic flux passes through the object, depending on the size and position of the error in the section, there will be a deflection and concentration of magnetic fluxes below it.

332

L. Radovanovi´c et al.

The different concentrations of magnetic fluxes affect the magnetic field changes on the object’s surface. When the test object is exposed to the effect of a magnetic field, field defects are formed in the places of magnetic discontinuities. Magnetic tests are based on the magnetic field’s dissipation above the object’s defect site. In a homogeneous object, the magnetic fluxes are straight lines. When the currents encounter a defect, they bend around it, forming a scattering magnetic field. (Fig. 1.) The size of the dispersion and deflection of the currents depend on the dimensions and depth of the defect and the direction of propagation of the defect in relation to the magnetic currents. The largest deviation occurs when the fault is normal to the direction of the magnetic fluxes. In order to correctly determine the size of the error, in practical conditions the test is performed in two mutually normal directions.

Fig.1. The spread of the magnetic field [1].

Magnetic particles are applied to the examined surface in a dry state or in a suspension (water or oil). The indicator of defects is the accumulation of ferromagnetic particles applied to the tested surface. Given that magnetic fluxes are invisible, and to make the error visible, the object’s surface is sprinkled with ferromagnetic powder mixed with transformer oil or kerosene. Magnetic powder is black magnetite Fe2O3. In some cases, colored and fluorescent magnetic suspensions are taken. Its luminous particles facilitate control in hard-to-reach places. The method can be used to measure: layer thickness, structure variations, grain size, hardness, etc. The principle of the method consists in the fact that magnetic currents must pass through the test object, which will deform around the defect in the material. This is shown in several ways: • Using the difference in voltage that is sent to the amplifier and the signal is displayed on the galvanometer. This principle is applied when testing welds, steel ropes, pipes, etc. • The test object is magnetized and coated with a rare oil with powdered iron (or by immersing it in a bath with such oil). Iron dust collects more intensively in places with an error.

Preventive Maintenance of Hydraulic Press Using Magnetic Testing

333

Devices used for magnetic tests are called ferroflux or magnetoflux devices. After the test, the magnetized parts must be demagnetized. Demagnetization of objects is done in special devices or a special device for demagnetization is built into the ferroflux device [7].

3 Advantages and Limitations of Magnetic Tests The main area of application of this method is in the examination of semi-finished products: metal rods and billets, sheets, forgings, and cast pieces. This method is also used for preventive maintenance: shafts, frames, supports, flywheels, hooks, turbine blades, etc. It detects well small and shallow surface cracks and errors under the surface that could be at a greater depth. This method will detect if the error is fine, sharp, and close to the surface (e.g., a non-metallic inclusion). The indication will be less noticeable if the error is at a greater depth. Indications appear directly on the examined surface and are reflected as images of the actual form of the error. There are practically no restrictions on the size or shape of the examined objects. No pre-cleaning is required. It is possible to detect even cracks that are filled with other materials. The limitations of the method are as follows: • • • • • •

Application solely for ferromagnetic materials. The magnetic field must be directed towards the intersections of the main error plane. Subsequent demagnetization is often required. Cleaning is required to remove magnetic particles. Large products require high currents. Care should be taken regarding local heating or burning of finished products or surfaces at electrical contact points. • Experience is required to interpret the indications. • The main disadvantage of the method: the possibility of determining the dimensions of errors. Influential factors to consider when establishing a magnetic particle testing procedure are: • • • •

Type of magnetic particles, Method of magnetization, Direction of magnetization, Electrical current (depending on the hardness of the material and the size of the defects), • Equipment, • Types of electrical current. Figure 2 shows the schematic diagram of flux leakage in the presence of geometric discontinuity. The material can be magnetized with two magnetic fields: longitudinal and circular. A discontinuity cannot be located if it is located longitudinally about the flow of the

334

L. Radovanovi´c et al.

Fig.2. Schematic diagram of flux leakage in the presence of geometric discontinuity [8].

magnetic field, i.e., parallel to the lines, because the discontinuity located in this way moves the magnetic lines very little. This means that it is necessary to ensure that the magnetic field flows as comfortably as possible on the discontinuity so that the discontinuity disrupts the field strength as much as possible. Therefore, this problem is solved by mutually alternating magnetization or circular flow of the magnetic field. It should be noted that peripherally located discontinuities will not be detected by a circular field, but most longitudinal and transverse errors, as well as those oriented at 45°, will. Materials can be magnetized, i.e., create a magnetic field in them in two ways: • Using direct induction • Using indirect induction. The material can be magnetized by passing an electric current through the tested material, which creates electromagnetic induction. Care must be taken to ensure that the tested material is not damaged due to heating in places of increased resistance to the flow of electric current or due to poor contact between the test equipment and the tested material. Direct induction can be achieved by inserting the tested material between two contacts through which an electric current is released, which results in the creation of a circular magnetic field according to the right-hand rule; if the electric current has the direction indicated by the thumb, the magnetic field around the conductors will flow in the direction the fingers of the right hand are pointing. Indirect induction is based on the creation of an external magnetic field that will magnetize the examined ferromagnetic material. The first and most widespread way of achieving indirect induction is by passing an electric current through the yoke, which creates a strong and controlled magnetic field that lasts only as long as electric current flows through the wire - a coil wrapped around a core of soft ferromagnetic iron placed inside the yoke. Applying the yoke under the action of electric current to the tested material creates a longitudinal magnetic field. There is no danger of heating the tested part because there is no flow of current through the tested part. Yoke magnetization is a very popular method due to its portability and low cost. Most yokes can work with both direct current and alternating current. The magnetic field inside the material can also be obtained by the electrodes so that the material is placed so that the current flows from one electrode to the other, creating a magnetic field around the path of the flow of electric current between the electrodes.

Preventive Maintenance of Hydraulic Press Using Magnetic Testing

335

The use of coils is the third method of achieving a magnetic field by indirect induction. The tested material is placed longitudinally in a concentric magnetic field inside a coil through which a electric current passes, creating a longitudinal magnetic field in the tested material. The use of permanent magnets is limited due to the impossibility of controlling the strength of the magnetic field. 3.1 Magnetizing Currents The currents used to magnetize the tested material using the previously mentioned procedures can play an important role in testing with magnetic particles. Therefore, the currents can be modified depending on the test requirements. The main classification of magnetizing currents is on two types, direct currents, and alternating currents. DC Current A direct current is a current in which the direction of the electric field does not change. If, in addition, the strength does not change, and therefore neither does the voltage, it is called constant direct current. Direct current sources in practical application are often galvanic elements, i.e., different types of batteries and accumulators. The direction of direct current is equal to the direction of the electric field, so it is opposite to the direction of electron movement. It is highly desirable for testing with magnetic particles because it can create a magnetic field that penetrates deep into the material [9]. Alternating Current Alternating current is often used for testing because it is practical and available in almost all areas where testing is performed. Alternating current, unlike the direct current, has a magnetic flux and a density of forces in time. However, the distribution of forces in space remains the same as with constant direct current. It is advisable to use alternating current if only surface errors are expected because the magnetic field produced by alternating current is concentrated on a narrow surface part of the tested material. The reason is that the phenomenon of alternating current means that its flow is concentrated on the surface of the conductors [10].

4 Testing Procedure on the Hydraulic Press Before the test, it is very important to determine what errors on the material are expected during the test and at what depths since the characteristics of the magnetic field that will be applied and the equipment that will produce the field with the required characteristics will depend on this. It is necessary to approach the selection of the device that will be used for magnetization, select the electric current, particles for indication, and equipment for illuminating the resulting accumulated particles. Magnetic particle testing equipment is divided according to the degree of portability [11]. Data on the object and examination procedure are presented in the following table (Table 1.) [12]. Figure 3 shows a device for testing cracks and ferromagnetic surfaces.

336

L. Radovanovi´c et al. Table 1. Data on the object and examination procedure.

Subject of inquiry Hydraulic press -Lipostroy HPT-1–630 Data on the object of the examination Label

HPT-1–630

Dimensions mm

3640.0x 1450.0 × 4020.0

Material

/

Drawing no

/

Welding procedure

111

Connection type

FW

Thermal treatment

/

Operating temperature, pressure

/

Time of exploitation

30–40 years

Scope of the examination

Estimated critical points

Data on the examination procedure Procedure / Instruction

MT U -S

Device type/mark/manufacturer

TIEDE Germany

Method

SRPS EN ISO 17638:2017

The distance between the poles

130 mm

Examination technique

Wet, black-white

Ambient temperature

12 C

Magnetization technique

EM yoke

Surface preparation

Grinding

Magnetic field strength

2kA/m

Lighting

Flashlight

Magnetizing current

Alternate Test side External

Alternate Test side External

Alternate Test side External

Demagnetization

/

Accessories

Loupe, meter

Quality Level ISO 5817

“C” Acceptability level SRPS EN ISO 23278 “3”

Acceptability criterion Quality Level ISO 5817

“C” Acceptability Level SEPS EN ISO 23278 “3”

Figures 4, 5, 6, 7, 8, 9, 10 show the cracks found and damaged on the examined hydraulic press. Figure 11 shows the base of the hydraulic press - Lipostroj HPT-1–630 (Fig. 9) and a sketch of its moving part (Fig. 12).

Preventive Maintenance of Hydraulic Press Using Magnetic Testing

Fig.3. TWM device for testing

Fig.4. Tested sample 1/1–1/2.

Fig.5. Tested sample 2/1–2/2.

337

338

L. Radovanovi´c et al.

Fig.6. Tested sample 2/2–2/3.

Fig.7. Tested sample 2/2–2/3.

Preventive Maintenance of Hydraulic Press Using Magnetic Testing

Fig.8. Tested sample 3/2–3/3.

Fig.9. Tested sample A2.

Fig.10. Tested sample B4.

339

340

L. Radovanovi´c et al.

Fig.11. Base of the press.

Fig.12. The moving part.

5 Conclusion Examination of the welded joints on the Litrostroy 630–1 HTP press, in the positions that were evaluated as critical places by the magnetic particle method, indications of the type of cracks were observed, which are unacceptable. In the “T” positions of the double-sided welded corner joints, where the middle part was not tested due to the working conditions, cracks appeared. At certain joints of corner joints, cracks came to the surface and could be confirmed along the entire vertical, while at other positions, the crack was observed only at the top. If it is necessary to determine the presence of inhomogeneity of the joint within the weld, it is possible to detect the defects using the ultrasonic test method. This method uses the echo of a transmitted pulse to determine the homogeneity of the material. The ultrasonic wave passes through the tested material and is reflected when it encounters the boundary surface between two environments of different acoustic resistance. If there are no errors in the material (part of the system), the ultrasonic waves will be reflected from the opposite surface of the piece, i.e., they will not be able to pass into the air environment. Based on the results of the magnetic test on the press in this paper, it can be concluded that certain samples passed the tests while a significantly larger number did not. Research into the causes of malfunctions is one of the important tasks of diagnosing the condition of technical systems, and it is carried out with the aim of pointing out the

Preventive Maintenance of Hydraulic Press Using Magnetic Testing

341

places and causes of malfunctions in the technical system. Non-destructive methods are often used in technical diagnostics. Given that technical diagnostics is part of preventive maintenance, i.e., it includes activities before failure, we can conclude that the application of non-destructive methods, i.e., magnetic tests, is very important in the improvement of preventive maintenance.

References 1. Adamovic, Ž.: Tehniˇcka dijagnostika. Tehniˇcki fakultet “Mihajlo Pupin”, Zrenjanin (2010) 2. Sacarea, A.I., Oancea, G., Parv, L.: Magnetic particle inspection optimization solution within the frame of NDT 4.0. Processes 9(6), 1067 (2021) 3. Goljat, S., Tiˇc, V., Lovrec, D.: Prevention of water ingress in hydraulic systems. Appl. Eng. Lett. 6(3), 99–104 (2021). https://doi.org/10.18485/aeletters.2021.6.3.2 4. Li, L., Huang, H., Zhao, F., Triebe, M.J., Liu, Z.: Analysis of a novel energy-efficient system with double-actuator for hydraulic press. Mechatronics 47, 77–87 (2017) - c, L., Kavali´c, M.: Analysis 5. Novakovi´c, B., Radovanovi´c, L., Zuber, N., Radosav, D., Ðordevi´ of the influence of hydraulic fluid quality on external gear pump performance. Eksploatacja i Niezawodno´sc´ 24(2) (2022). https://doi.org/10.17531/ein.2022.2.7 6. Todi´c, N., Savi´c, S., Jovanovi´c, S., Djordjevi´c, Z.: Research of water hydraulic components and systems from aspects of quality of life. Appl. Eng. Lett. 4(3), 93–97 (2019). https://doi. org/10.18485/aeletters.2019.4.3.3 7. https://repozitorij.vuka.hr/islandora/object/vuka%3A310/datastream/PDF/view. Accessed 01 July 2023 8. Wang, Z.D., Gu, Y., Wang, Y.S.: A review of three magnetic NDT technologies. J. Magn. Magn. Mater. 324(4), 382–388 (2012) 9. Essert, M., Grilec, J.: Elektricitet i magnetizam - fizikalne osnove, Digitalniudžbenik, Fakultet strojarstva i brodogradnje, Zagreb (2009) 10. https://www.fsb.unizg.hr/usb_frontend/files/1348657703-0-eim_2009.pdf. Accessed 01 July 2023 11. Pinter, V.: Osnove elektrotehnike - Knjiga prva, Tehniˇcka knjiga D.D., Zagreb (1994) 12. http://www.idef.hr/ndt-magnetna-metoda. Accessed 01 July 2023

Diagnostics of Journal Fluid-Film Bearing Failure Using a Data Manager System – Case History Marko Katini´c1(B) , Pejo Konjati´c1 , Mirko Karakaši´c1 , and Danko Glavaš2 1 Mechanical Engineering Faculty, University of Slavonski Brod, Slavonski Brod, Croatia

[email protected] 2 Petrokemija, Plc. Fertilizer Company, Kutina, Croatia

Abstract. The journal fluid-film bearing is a vital part of a rotary machine, which has a high load-carrying capacity. Babbitt bearings are commonly used to support the shafts of high-speed machines. Severe damage to the babbitt layer can lead to damage of the shaft journal. Fluid-film bearings installed in high-speed machines are exposed to damage due to the action of various mechanisms. It is very important to diagnose bearing damage at an early phase to prevent the possible failure of the machine. A machine monitoring system of shaft vibrations and bearing temperatures is very useful for early diagnosis of machine bearing condition. In this paper, a practical example of diagnostics of journal fluid-film bearing failure in a gear unit using a specialized data manager system is given. The bearing damage was confirmed by inspections performed during the downtime of the gear unit. Based on the visual investigation, the most likely primary cause of bearing damage was electrical pitting. Keywords: Journal Fluid-film Bearing · Gear Unit · Failure · Data Manager System · Electrical Pitting

1 Introduction Fluid-film bearings are vital elements of machines that allow shaft guidance and load transfer from the shaft to the bearing support. The sliding surfaces of the bearings are usually continuously lubricated with oil. Babbitt bearings are commonly used to support the shafts of machines. Thrust and journal fluid-film bearings are distinguished according to the direction of load action. The journal fluid-film bearing transmits static and dynamic loads acting perpendicular to the shaft journal axis of the machine. On the other hand, the thrust fluid-film bearing transmits the axial load. The vast majority of fluid-film bearings are in long-term reliable operation. However, damage to the fluid-film bearing can often lead to very costly consequences, such as accidental damage to the machine and production downtime. Fluid-film bearings installed in high-speed machines are subject to different modes of damage caused by various actions. The cause of the damage may be obscure. Damage © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 342–350, 2024. https://doi.org/10.1007/978-3-031-51494-4_28

Diagnostics of Journal Fluid-Film Bearing Failure

343

is often the result of a combination of several failure modes. Sometimes the bearings themselves are blamed for the damage. However, the cause of damage is more frequently external to the bearing than in its design or manufacture. There are many possible causes of bearing damage such as: bearing bore distortion, black scab or wire wool damage, cavitation erosion, corrosion, electrical pitting, fatigue, fretting, lining extrusion and cracking, pivot fatigue, scoring, spark tracking, thermal faceting, uneven wear, wiping, etc. [1–3]. A damaged bearing usually leads to unplanned downtime of a vital process plant machine, such as a steam turbine, compressor, pump, and gear unit. The consequence is the interruption of the production process, which represents significant economic losses, and sometimes, unfortunately, can endanger the health and lives of process personnel. In the petrochemical industry, these vital machines are often located in explosively hazardous areas, which further warns of the seriousness of this issue. To protect the vital elements of the machine as effectively as possible (rotor, gear or similar), it is necessary to monitor the condition of the fluid-film bearings during the operation of the machine. This involves monitoring bearing temperatures and vibrations, as these are the two most important indicators for predicting potential bearing damage. To monitor these parameters, a specialized monitoring system and a data manager system for data acquisition and processing are required. This paper presents a practical example of successful diagnostics of journal fluid-film bearing damage in a gear unit using a specialized data manager system. The detected bearing damage was confirmed by controls performed during the downtime of the gear unit. Based on the inspection, the most likely primary cause of bearing damage was the electrical pitting.

2 Technical Data of the Gear Unit The CO2 compressor train is a vital part of the urea production plant, and consists of a steam turbine, as a drive machine, and a low-pressure (LP) compressor, single-stage speed increaser gear unit and high-pressure (HP) compressor (Fig. 1).

Fig. 1. The CO2 compressor train.

Figure 2 shows a cross-sectional drawing of the single-stage gear unit, located between the LP and the HP centrifugal compressor. Cylindrical helical gears are connected to the compressor rotors by gear couplings. The gears are supported in two-piece

344

M. Katini´c et al.

journal fluid-film bearings. The axial force of each gear is carried by the corresponding tilting pad thrust bearing.

Fig. 2. Cross-sectional drawing of the single-stage gear unit [4].

The basic technical data of the gearbox unit are as follows: – – – – – –

Rated power: 4760 kW Gear ratio: 1,942 Number of drive gear teeth: 101 Number of driven gear teeth: 52 Maximum rotation speed of the drive gear: 14884 rpm Maximum rotation speed of the driven gear: 7560 rpm

3 Monitoring System and Data Manager System The shaft vibrations, bearing temperatures and axial displacements of the rotors and gears are continuously monitored by the Bently Nevada 3500 monitoring system. The shaft vibrations and shaft centerline positions are monitored by proximity probes built into the compressor bearing housings or gear unit casing. Figure 3 shows the measurement scheme of the gear unit.

Diagnostics of Journal Fluid-Film Bearing Failure

345

Fig. 3. Measurement scheme of the gear unit.

Proximity transducers (probes) for measuring vibrations operate on the principle of eddy currents and these transducers are preferred for monitoring the vibrations of machines with fluid-film bearings. Measurement of shaft vibrations is usually performed using two orthogonal, coplanar proximity transducers. The probes are mounted at 45° left and 45° right (Fig. 3). In addition to measuring relative shaft vibrations (alternating signal components), the proximity probes also measure the average position of the shaft (DC signal component). Changes in shaft position can provide very important diagnostic information [4]. Proximity probes must be set at an appropriate distance (DC distance) from the target shaft surface during their installation. In the case of the considered gear unit, the DC gap of the probe tip towards the shaft surface has been set to –10 VDC (physical gap about 1.25 mm). Each probe has a linear characteristic in the physical gap range of 0.25 mm to 2 mm. If the distance of the shaft from the tip of the probe decreases, the output voltage is less negative and vice versa. The Bently Nevada Data Manager 2000 system is a computer system used for online data acquisition (vibration, displacements, temperatures), archiving and display in suitable formats. Data Manager 2000 system connects to the 3500-monitoring system through communication processors. This system collects static and dynamic data under both steady-state and transient operating conditions of the machine. Data Manager 2000 stores live and historical data, and can present the data in several different formats, including bar graph, waveform (orbit, timebase, etc.), trend (live, historical, high resolution, etc.) and spectral (synchronous, asynchronous, and full).

346

M. Katini´c et al.

4 Diagnostics of Journal Fluid-Film Bearing Failure As already explained, shaft position changes can provide very important diagnostic information [5]. The average shaft centerline plot, taken from the Data Manager 2000 system, provides this information. This plot is designed to show changes in the average position of the shaft; thus, the plot is effectively low-pass filtered and does not display rapidly changing (dynamic) data [5]. Figure 4 shows the average shaft centerline plot of the drive gear in the corresponding journal bearing in the period from June 1 to August 10, 2021. It is clearly visible that the shaft journal has moved discontinuously down to the left looking away from the drive machine. It is also clear that the shaft centerline went beyond the set limit of the bearing clearance, which indicates the intensive wear of the bearing babbitt and increase radial clearance. The estimated increase in clearance from the available data is approximately 0,15 mm.

Fig. 4. Average shaft centerline plot.

These changes in the position of the shaft centerline can be also seen from the trends of gaps on both probes. Figure 5 shows the trend of gap change on the XE106–2 probe. There were several machine outages during this period. As can be seen the machine was driven in the following periods: from 19.6. to 22.6., from 28.6. to 7.7. And from 8.7. to 10.8. The DC gap on this probe was stable until 8.7. From 8.7. The value of the gap began to change and became more and more negative. These gap changes were not continuous. Namely, at some moments there were step changes of the gap of the order of 0,1 to 0,2 V. When the gap reached -11V, the machine was stopped to avoid catastrophic damage. A significant increase in clearance causes misalignment of gears in mesh and poorer tooth contact pattern, and thus the variation of the stress distribution in the contact region and the tooth root. Equivalent stresses are directly proportional with the misalignment deviation [6]. The stress concentration is higher at the contact region and the tooth root with augmentation of misalignment angle [6]. The journal bearing temperature during the operation of the machine was quite stable, with no significant changes (Fig. 6).

Diagnostics of Journal Fluid-Film Bearing Failure

347

Fig. 5. The time trend of the gap on probe 22XE106–2.

Fig. 6. The time trend of the journal bearing temperature.

5 Fact Finding After Disassembling the Machine After disassembling the machine, a visual inspection revealed damage to the journal bearing and the associated gear journal. Figure 7 shows the appearance of the babbitt surface of the bearing halves. The surface of the loaded half of the bearing appears to be partially frosted (grey). The rest of the surface is shiny. The surface of the gear journal is shown in Fig. 8. Also, the journal surface looks like it is frosted. The results of measuring the inner diameter of the bearing and the outer diameter of the gear journal are given in Table 1. It is clear that both the bearing and the journal are unevenly worn. The result is increased and uneven clearance in the bearing. The value of the design clearance in the bearing should be within the range of 0,16 mm to 0,19 mm. The measured clearance in the middle of the bearing (in the vertical direction) was 0,31 mm.

348

M. Katini´c et al.

Fig. 7. Frosted half of the journal bearing.

Fig. 8. Frosted gear journal.

By visual comparison the damaged bearing and journal surfaces with different examples of bearing damage from the literature [3, 7], it was preliminarily concluded that the probable primary cause of the damages is electrical pitting. Electrical discharge through the oil film between the shaft and the bearing in non-electric rotary machines (steam turbines, compressors, gear units, etc.) can occur due to the build-up of static electricity. The gear unit is located in a potentially explosive zone and therefore there are no built-in brushes to remove static electricity. Because of this, there is a high probability of static electricity in this machine. The electrical discharge can cause severe pitting of the bearing and shaft journal surfaces. In extreme cases, damage can occur very rapidly

Diagnostics of Journal Fluid-Film Bearing Failure

349

Table 1. Dimensional control of bearing inner diameter and gear journal outer diameter.

Measurement sketch

Measured diameters a = 90,140 mm b = 90,065 mm c = 90,075 mm

a = 89,845 mm b = 89,830 mm c = 89,840 mm

[7]. When it does, the cause may be difficult to diagnose, as pitting of the bearing surface is followed ultimately by wiping, which can obscure the original pitting [7]. To confirm the assumed cause of the damage, it would be necessary to carry out additional analyzes of the surfaces of the bearing and the journal of the gear, which would include metallographic tests and tests with a scanning electron microscope.

6 Conclusions It is very important to detect machine malfunction at an early stage of its development to avoid catastrophic failure. The vibration monitoring system and data manager system are very useful in machine diagnostics. In a practical example, the detection of fluid-film journal bearing wear using the data manager system is presented. In this case, the average shaft centerline plot was useful. This plot produced a very powerful and complete picture of the response of the shaft journal relative to the bearing. Evidently, the plot detected the bearing wear. This was also confirmed by visual and dimensional inspection of the bearing during machine downtime. It would be useful to install an additional system for continuous monitoring and protection of the machine’s bearings against the effect of electric discharge.

References 1. Conway-Jones, J.M., Leopard, A.J.: Plain bearing damage. In: The Proceedings of the 4th Turbomachinery Symposium, Texas A&M University, Texas (1976)

350

M. Katini´c et al.

2. Zeidan, F.Y., Herbage, B.S.: Fluid film bearing fundamentals and failure analysis. In: The Proceedings of the 20th Turbomachinery Symposium, Texas A&M University, Texas (1991) 3. Blair, B.J., Pethybridge, G.: Hydrodynamic bearing damage and remediation of contributing factors in rotating machinery. In: 9th EDF/Pprime (LMS) Poitiers Workshop (2010) 4. Technical documentation from Petrokemija d.d. Kutina 5. Bently, D.: Fundamentals of Rotating Machinery Diagnostics, 1st edn. Bently Pressurized Bearing Press, Minden (2002) 6. Lias, M.R., Rao, T.V.V.L.N., Awang, M., Khan, M.: The stress distribution of gear tooth due to axial misalignment condition. J. Appl. Sci. 12(23), 2404–2410 (2012) 7. Bearing damage index. https://www.waukbearing.com/en/resources/bearing-damage-index/ electrical-pitting.html. Accessed 29 Aug 2023

Predicting the Specific Gravity of Must During Fermentation Using Machine Learning Models Ivana Kovaˇcevi´c, Mihaela Ori´c, Ivana Hartmann Toli´c(B) and Emmanuel Karlo Nyarko

,

Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Croatia Kneza Trpimira 2B, 31000 Osijek, Croatia [email protected]

Abstract. In this article, the process of fermentation in winemaking is investigated. The authors analyze fermentation data and propose several machine learning models to assess the effectiveness of ongoing fermentation processes based on measurements of specific gravity and temperature. To evaluate the performance of each model, various metrics such as R-squared, Mean Squared Error, Root Mean Squared Error, and Maximum Absolute Error are employed. The results suggest that the models have the potential to predict fermentation outcomes in winemaking and provide winemakers with a valuable tool to optimize the process and ensure high-quality wine production. Keywords: Fermentation · Machine Learning · Specific gravity · Winemaking

1 Introduction Spontaneous fermentation usually produces traditional fermented foods, depending on climatic conditions [1]. The wine production process includes multiple separate production stages, similar to most other foods: harvesting, crushing and pressing, fermentation, ageing and maturation, clarifying and stabilization, bottling, and, eventually, distribution and consumption. These stages in the winemaking process require an artful blending of science and technique, and each one affects the wine’s final qualities, traits, and enjoyment. Fermentation is the most vital and significant stage in winemaking, where yeast converts sugars in the must into alcohol, carbon dioxide, and other compounds. Monitoring the wine fermentation process is crucial for producing high-quality products every winemaker wants. For regulating and observing fermentation, temperature and specific gravity are essential variables. While specific gravity offers information on sugar concentration and possible alcohol generation, temperature affects yeast metabolism. For yeast activity and the successful completion of the fermentation process, it is essential to maintain the ideal fermentation temperature. Temperature affects yeast growth rate, sugar use, and the formation of aroma and flavour compounds [2]. Excessive yeast metabolism can occur at high temperatures, resulting in the generation of unpleasant off-flavours and off-odours. Low temperatures, on the other hand, can impede fermentation or cause yeast dormancy. Winemakers employ various techniques, such as © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 351–363, 2024. https://doi.org/10.1007/978-3-031-51494-4_29

352

I. Kovaˇcevi´c et al.

cooling jackets, temperature-controlled tanks, or refrigeration systems, to regulate and maintain the desired fermentation temperature range [3] and control product quality [1]. A valuable metric for determining the sugar content and probable alcohol levels in the must is specific gravity. It is typically measured using hydrometers or refractometers. Specific gravity readings are taken before and during fermentation, providing insights into the progress and completeness of fermentation. Due to the must’s high sugar content, the specific gravity is initially high. The specific gravity continuously decreases as yeast consumes sugar and produces alcohol. Monitoring specific gravity enables winemakers to calculate the wine’s potential alcohol content and determine the end of the fermentation process. The authors of this article are involved in a project to develop an expert system to help winemakers in the production of wine. As mentioned earlier, one of the steps in wine production is fermentation. In this paper, the authors analyze data from a local Croatian winery through the successful fermentation of three different white wines. Based on this data, the authors propose three preliminary models that could be used to determine whether the ongoing fermentation process is optimal or not.

2 Methodology in Related Research Machine learning techniques (ML) are gaining popularity and are used in many fields to detect and classify patterns in complex data structures. ML algorithms are applicable in various real-world application areas such as IoT systems, cybersecurity services, business and recommendation systems, smart cities, healthcare, context-aware systems, sustainable agriculture, computer vision, natural language processing, and many more. The selection of an appropriate ML method depends on factors such as the type of data, the availability of labeled data, and the specific problem or task. The production of wine involves a biochemical process known as alcoholic fermentation. This process is very complex, as yeast releases a variety of metabolites at different concentrations, which are closely related to the physicochemical and sensory properties of the resulting wine [4]. Extensive research has been carried out to thoroughly analyze, model, and control each stage of winemaking, with the aim of streamlining the management process. Henriques et al. [5] propose a methodology for modeling cold fermentations with non-conventional Saccharomyces species by establishing an experimental modeling pipeline. This pipeline involves conducting small-scale micro vinifications and wine fermentations under controlled conditions. During these fermentations, the authors monitor growth rate and critical extracellular metabolites such as glucose, fructose, ethanol, glycerol, and acetic acid. They compare several candidate models that account for different biomass growth, transport, and inhibition mechanisms described in the literature. Developing a minimal model with the fewest required parameters ensures structural identifiability, allowing the model to be uniquely fit to the data while improving practical identifiability. For the most successful models, a combined modeling strategy is used to increase robustness and minimize predictive uncertainty. Results are presented quantitatively and an ensemble of models is used to make robust predictions for processing conditions, such as initial inoculation and temperature, to optimize alcohol and glycerol production.

Predicting the Specific Gravity of Must During Fermentation

353

In [6], the authors combine in situ density measurements based on differential pressure with a kinetic wine model and parameter estimation routine to predict the progress of a 1500 l wine fermentation. Wei et al. in [7] study the volatile organic compounds (VOCs) profile of pomelo wine and develop an index to predict the degree of fermentation, leading to improved quality in industrial production. They use gas chromatography-mass spectrometry to analyze the changes in VOC content during different fermentation periods. Principal component analysis, cluster analysis and partial least squares (PLS) regression are used to evaluate the variations of VOC content during different fermentation phases. PLS models identify the α-phellandrene/geraniol ratio as a potential index for determining the degree of fermentation of pomelo wine. Multivariate analysis techniques are also used to create an index to predict the degree of fermentation of pomelo wine under controlled laboratory conditions. The proposed index provides winemakers and wine chemists with valuable insights into the VOC composition and profile of pomelo wine during the fermentation process. Tardaguila et al. in [8] provide an overview of smart applications and digital technologies currently used in viticulture. Wine quality evaluation involves sophisticated and diverse testing methods. Advice from a wine expert, while valuable, can be expensive and subjective. Therefore, several machine learning-based methods for wine quality prediction have been proposed, such as authors in [9] and [10]. The integration of artificial intelligence (AI) and machine learning techniques in fermentation bioprocesses is proposed in [11]. They combine the fields of computer engineering and food engineering to highlight the importance of digitalization in modern industrial processes. Their research involves the development and implementation of a software application that models a fully configurable three-layer feed-forward multilayer perceptron neural network. This neural network, trained with experimental data, can predict the evolution of quantities characteristic of the alcoholic fermentation process in white wines. Román et al. [12] propose the use of artificial neural networks (ANNs) to effectively identify nonlinear patterns in wine fermentation processes and accurately predict fermentation problems. They investigate different scenarios by varying the prediction variables and the duration of fermentation. Predicted variables include total sugars, alcohol, glycerol, density, organic acids and nitrogen compounds. The study highlights the potential of ANNs as valuable tools for monitoring and optimizing wine production processes. Urtubia et al. [13] examine two cases related to the prediction of problematic wine fermentations. In the first case, they employ the Support Vector Machine (SVM) algorithm to analyze different groups of chemical variables, including amino acids, saturated fatty acids, unsaturated fatty acids, organic acids, and fermentation control variables. In the second case, they examine individual chemical variables, namely density, yeast assimilable nitrogen (YAN), sugar content (Brix), and acidity, as predictors of problematic wine fermentation using SVM.

354

I. Kovaˇcevi´c et al.

3 Wine Fermentation Data The dataset used for the analysis includes fermentation data collected by a Croatian winery over an extensive eleven-year period from 2010 to 2020. The winery uses measurements of the temperature and specific gravity to implement effective control strategies throughout the fermentation process. Temperature control involves continuous monitoring and adjustment of fermentation temperature to create an ideal environment for yeast activity. Depending on the wine style and yeast strain used, this can be done by cooling or heating. Specific gravity measurement is used to help make decisions about adding nutrients, adjusting sugar levels, or determining the optimal end point for fermentation. The dataset provided includes recorded series of fermentations, with each data sample containing two measurements: specific gravity (γ ) in units of Oechsle [°Oe] and temperature (T ) in degrees Celsius [°C]. The unit Oechsle [°Oe] is commonly used in winemaking to estimate the potential alcohol content of grapes or grape must. It is based on the specific gravity of the juice or must and reflects the soluble solids, mainly sugars, present in the juice. This sugar correlates directly with the potential alcohol that can be produced during fermentation. The Oechsle scale usually ranges from 50°Oe to 200°Oe, with higher values indicating a higher sugar concentration and potential alcohol content. It is important to note that the Oechsle unit is used primarily in German winemaking and plays an important role in classifying German wines according to ripeness and sugar content. Other wine regions may use other scales and units, such as Brix in the United States or Baumé in France, to measure sugar content and potential alcohol content. The data set contains only information from successful fermentations of three different white wines, namely Graševina, Chardonnay and Pinot Gris. The distribution of the measured data is shown in Fig. 1, and a comprehensive summary can be found in Table 1. Table 1. Total number of data samples per grape variety Graševina

Chardonnay

Pinot Gris

Total no. of fermentation processes

20

22

22

Total no. of data samples

397

393

329

A total of 64 different fermentation processes were recorded, comprising of 1119 data samples. The minimum and maximum readings, and average values of the data samples are shown in Table 2. It can be seen that the specific gravity values measured ranged from 26°Oe to 104°Oe, while the temperature values ranged from 12.4 °C to 24 °C, with average values of 53.73°Oe and 17.19 °C (Table 2). The data in Table 3 shows that during the fermentation process, the average temperature, as well as specific gravity, of Chardonnay is slightly higher compared to Graševina and Pinot Gris. The number of data samples recorded only per specific gravity and per grape variety is shown in Fig. 2.

Predicting the Specific Gravity of Must During Fermentation

355

Fig. 1. Number of fermentation processes recorded per year and per grape variety

Table 2. Minimum, maximum and average values of the data measured during fermenations. Parameter Measured

Min. Value

Max. Value

Average value

Specific Gravity, γ [°Oe]

26

104

53.73

Temperature T [°C]

12.5

24

17.19

Interval between measurements [days]

1

5

-

No. of data samples per fermentation process

7

34

-

Table 3. Average measured Temperature T and average Specific Gravity γ per grape variety Grape variety

Parameter Measured Specific Gravity, γ [°Oe]

Temperature T [°C]

Graševina

51.59

16.47

Chardonnay

55.67

18.33

Pinot Gris

53.99

16.69

An example of a fermentation process is shown in Fig. 3. It can be seen that during the fermentation process, the specific gravity value decreases with time, indicating a decrease in the sugar content, and an increase in the alcohol content. To model the fermentation process, the authors used the recorded data and divided it into two subsets: the training set and the test set. The distribution of the measured data among these subsets is shown in Fig. 4, which shows a distribution ratio of 20:80, based on the total number of fermentation processes and the total number of data samples. The

356

I. Kovaˇcevi´c et al.

Fig. 2. Number of data samples recorded per specific gravity and per grape variety

Fig. 3. Example of a fermentation process showing the specific gravity and corresponding temperature.

training set was used to determine the model parameters, while the test set was used to evaluate the performance of the model. It is important to note that the division of the data was done without considering the grape variety. However, it is critical that a complete fermentation process or series be included in either the training set or the test set.

Predicting the Specific Gravity of Must During Fermentation

357

4 Fermentation Modelling As mentioned in the previous section, the dataset consists of recorded series of fermentations, with each data sample containing two measurements: specific gravity (γ ) and temperature (T ). Available also is the time interval between measurements (d), mainly being one day, as well as the grape variety (gv ). Hence a data sample x can be considered to be a tuple of 5 elements x = (k, γ , T , d , gv ) where k = (0, 1, . . . ) denotes the index in the sequence of measurement or the sample number. In this paper, the model to be determined, F, is defined by the following equation: γ (k + 1) = F(T (k), γ (k), d (k), T (k + 1), gv , k)

(1)

Fig. 4. Distribution of the measured data into training set and test set

The model attempts to estimate the specific gravity at time k +1 for a given fermentation process, given all the measurements provided at time k, and based on the temperature at time k + 1, i.e., T (k + 1). Such a model would help a wine producer to control the fermentation process by providing an estimate of the specific gravity value after one day given the temperature of the must after one day. Hence, by adjusting T (k + 1), the producer will be able to estimate γ (k + 1). The model to be determined according to Eq. (1), F, has 6 input variables (features) and one output variable. To determine the relevance of each of the input features, recursive feature elimination (RFE) was applied to the training data. RFE is a wrapper-type feature selection algorithm that begins by searching for a subset of features, starting with all the features in the training dataset, and successfully removing features until the desired number remain. This is achieved by adapting a specific ML algorithm used in the core of the model, ranking the features in order of importance, discarding the least important features and adapting the model again. This process is repeated until a certain number of features remain. The features are evaluated either using ML model provided (e.g. using the random forest importance criterion) or using a more general approach, e.g. a statistical method that is independent of the ML model provided [15]. In this paper RFE

358

I. Kovaˇcevi´c et al.

Table 4. The rank of each of the input features determined by recursive feature elimination Input variable

Rank/Importance

T (k)

4

γ (k)

1

d (k)

6

T (k + 1)

2

gv

5

k

3

was implemented using a Random Forest regressor. The obtained rank or importance of each of the input features is displayed in Table 4. Analyzing Table 4, it can be concluded that the three most relevant input features are γ(k), T (k + 1) and k, ie., the specific gravity measured at time k, the temperature at time k + 1 and the index of the current sample. Hence, a new model is proposed based on only these three input features and is defined by the following equation: γ (k + 1) = F3 (γ (k), T (k + 1), k)

(2)

In this paper, three ML models are considered for F (Eq. 1) and F3 (Eq. 2) separately: a multivariate linear regression model, a random forest model and an artificial neural network model. All modelling was done in Python using ML library scikit-learn [15]. Before training the models, all data were preprocessed by performing data standardization, i.e. all input parameters were transformed to have a mean of 0 and a standard deviation of 1. 4.1 Evaluation of the Models All models considered in this paper are evaluated using the following evaluation metrics: R-squared (R2 ), Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and Maximum Absolute Error (MAE). R2 represents the proportion of the variance in the dependent variable that is explained by the independent variables. It ranges from 0 to 1, with higher values indicating better fit. It is defined by Eq. 3:  (yi − fi )2 (3) R2 = 1 −  i 2 i (yi − yM ) where: yi – represents the measured or actual value, fi – represents the predicted value, yM – represents the mean of the measured value. MSE (Eq. 4) measures the mean squared difference between predicted and actual values. It provides a general measure of the accuracy of the model, with lower values

Predicting the Specific Gravity of Must During Fermentation

359

indicating better performance. MSE =

N 1  (yi − fi )2 N

(4)

i=1

RMSE (Eq. 5) is the square root of MSE and provides a more interpretable metric in the same units as the dependent variable. √ RMSE = MSE (5) MAE (Eq. 6) is the maximum absolute difference between predicted and actual values. It gives an indication of the largest individual prediction error of the model. MAE = max(|yi − fi |)

(6)

4.2 Machine Learning Models For each of the proposed models, F (Eq. 1) and F 3 (Eq. 2), three ML models were determined: a multivariate linear regression model, a random forest model and an artificial neural network model. 4.2.1 Multivariate Linear Regression Model Based on the training data, multivariate linear regression models are determined for Eq. 1. (Eq. 7) and Eq. 2 (Eq. 8): γ (k + 1) = 50.138 + 0.378T (k) + 19.901γ (k) − 0.216d (k) −0.780T (k + 1) + 1.469gv − 0.203k

(7)

γ (k + 1) = 50.138 + 19.880γ (k) − 0.250T (k + 1) + 1.504k

(8)

The performances of the linear regression models on the test dataset are shown in Table 4. Analysing Table 5, it can be concluded that both multivariate linear regression models have a high R2 value of at least 0.9850 and a low RMSE value of at most 2.4229. It can also be concluded that the multivariate linear regression model involving only three input features (Eq. 8) is the better of the two models. Table 5. Performance metrics of the multivariate linear regression models on the test dataset Model

R2

MSE

RMSE

MAE

F→ (Eq. 7)

0.9850

5.8704

2.4229

7.3507

F3 → (Eq. 8)

0.9854

5.7129

2.3902

7.2288

360

I. Kovaˇcevi´c et al.

Fig. 5. Comparison of the output values of the multivariate linear regression model (Eq. 8) with the actual values

The output values of the multivariate linear regression model (Eq. 8) obtained using the test dataset is displayed in Fig. 5. Looking at Fig. 5, it can be concluded that the multivariate linear regression model models the fermentation process relatively well. However, it should be noted that it does not estimate the initial fermentation values very well. 4.2.2 Artificial Neural Network Model A simple feed-forward neural network model with a single hidden layer was used. The logistic-sigmoid activation function was used for the neurons in the hidden layer, while the linear activation function was used for the neuron in the output layer. The number of neurons in the hidden layer was determined using GridSearchCV function in the scikit-learn library. Depending on the model being determined, F (Eq. 1) or F3 (Eq. 2) the number of input neurons were 6 and 3 respectively. The performances of the artificial neural network models are shown in Table 6. Table 6. Performance metrics of the artificial neural network models on the test dataset Model

R2

MSE

RMSE

MAE

F(Eq. 1)

0.9880

4.6865

2.1648

7.3474

F3 (Eq. 2)

0.9885

4.5015

2.1217

6.6514

Predicting the Specific Gravity of Must During Fermentation

361

Table 6 shows that the artificial neural network models have a high R2 value of at least 0.9880 and a low RMSE value of at most 2.1648. From the two neural network models, it can be concluded that the 3-input neural network model is the better of the two models. The output values of the neural network model using only three inputs obtained using the test dataset is displayed in Fig. 6. Looking at Fig. 6, it can be concluded that the artificial neural network model represents the fermentation process quite well.

Fig. 6. Comparison of the output values of the 3-input artificial neural network model with the actual values

4.2.3 Random Forest Model Two random forest models were determined based on the training data for F (Eq. 1) and F3 (Eq. 2). The default parameters of the random forest models were used as defined in ML library scikit-learn: the number of trees in the forest = 100, there is no limit for the maximum depth of the trees, the minimum number of samples required to split an internal node = 2; the minimum number of samples required to be at a leaf node = 1. The performances of the random forest models are shown in in Table 7. Table 7. Performance metrics of the random forest models on the test dataset Model

R2

MSE

RMSE

F(Eq. 1)

0.9863

5.3740

2.3182

8.1100

F3 (Eq. 2)

0.9842

6.1848

2.4869

10.4950

MAE

362

I. Kovaˇcevi´c et al.

Table 7 shows that the random forest models have a high R2 value of at least 0.9842 and a low RMSE value of at most 2.24869. From the two random forest models, it can be concluded that the model with all 6 inputs is the better of the two models. The output values of the random forest model using all six inputs obtained using the test dataset is displayed in Fig. 7. Looking at Fig. 7, it can be concluded that the random forest model represents the fermentation process relatively well.

Fig. 7. Comparison of the output values of the random forest model (number of inputs = 6) with the actual values

The comparison of all three models shows that the random forest model (number of inputs = 6) is comparable to the multivariate linear regression model (number of inputs = 3), which is slightly better. The artificial neural network model (number of inputs = 3), on the other hand, slightly outperforms the multivariate linear regression model. The R2 value increases by 0.3% and the values of MAE and RMSE decrease by 8% and 11% respectively for the artificial neural network model compared to the multivariate linear regression model.

5 Conclusions The aim of this work was to develop models to evaluate the fermentation process of white wines based on specific gravity and temperature measurements. The models showed good performance, with the artificial neural network model having a slight advantage in terms of accuracy over the multivariate linear regression model, which was comparable to the random forest model. The multivariate linear regression model showed high accuracy in estimating the specific gravity at the next time step, as shown by the high R2 value of 0.9854 and the low RMSE value of 2.3902. However, the model had difficulty in

Predicting the Specific Gravity of Must During Fermentation

363

accurately estimating the values for the initial fermentation. In contrast, the artificial neural network model with a single hidden layer and three neurons provided a good representation of the fermentation process. These results suggest that the models have the potential to predict fermentation outcomes in winemaking and provide winemakers with a valuable tool to optimise the process and ensure high-quality wine production. The applicability of the models to different wine varieties and the investigation of additional factors that affect the fermentation process could both be the subject of future research. Acknowledgments. We thank Krauthaker Vineyards and Winery for providing the fermentation data used in this research. This work results from implementing research activities on the project Development of an expert system for food production and processing management (ESIAExpert System for Intelligent Agriculture KK.01.1.1.07.0036) funded by the European Regional Development Fund.

References 1. Liang, F., Ban, S., Huang, H., Che, F., Wu, Q., Xu, Y.: Predicting the effect of climatic factors on diversity of flavor compounds in Daqu fermentation. Lwt 169, 113984 (2022) 2. He, Y., et al.: Wort composition and its impact on the flavour-active higher alcohol and ester formation of beer–a review. J. Inst. Brew. 120(3), 157–163 (2014) 3. Butzke, C.E.: Winemaking Problems Solved. Elsevier (2010) 4. Schorn-Garcia, D., et al.: ATR-MIR spectroscopy as a process analytical technology in wine alcoholic fermentation - a tutorial. Microchem. J. 166, 106215 (2021) 5. Henriques, D., Alonso-del-Real, J., Querol, A., Balsa-Canto, E.: Saccharomyces cerevisiae and S. kudriavzevii synthetic wine fermentation performance dissected by predictive modeling. Front. Microbiol. 9(FEB), 317309 (2018). https://doi.org/10.3389/FMICB.2018.00088/ BIBTEX 6. Nelson, J., Boulton, R., Knoesen, A.: Automated density measurement with real-time predictive modeling of wine fermentations. IEEE Trans. Instrum. Meas. 71, 1–7 (2022) 7. Wei, Q., Liu, G., Zhang, C., Sun, J., Zhang, Y.: Identification of characteristic volatile compounds and prediction of fermentation degree of pomelo wine using partial least squares regression. LWT 154, 112830 (2022) 8. Tardaguila, J., Stoll, M., Gutiérrez, S., Proffitt, T., Diago, M.P.: Smart applications and digital technologies in viticulture: a review. Smart Agric. Technol. 1, 100005 (2021) 9. Chiu, T.H.Y., Wu, C.W., Chen, C.H.: A hybrid wine classification model for quality prediction. In: Del Bimbo, A., et al. (eds.) Pattern Recognition, ICPR 2021, vol. 12664, pp. 430–438. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-68799-1_31 10. Lee, C.K.H., Law, K.M.Y., Ip, A.W.H.: A rule-based quality analytics system for the global wine industry. J. Glob. Inf. Manage. (JGIM) 29(3), 256–273 (2021) 11. Florea, A., Sipos, A., Stoisor, M.C.: Applying AI tools for modeling, predicting and managing the white wine fermentation process. Fermentation 8(4), 137 (2022). https://doi.org/10.3390/ FERMENTATION8040137 12. Román, R.C., Hernández, O.G., Urtubia, U.A.: Prediction of problematic wine fermentations using artificial neural networks. Bioprocess Biosyst. Eng. 34, 1057–1065 (2011) 13. Urtubia, A., León, R., Vargas, M.: Identification of chemical markers to detect abnormal wine fermentation using support vector machines. Comput. Chem. Eng. 145, 107158 (2021) 14. Kuhn, M., Johnson, K.: Applied Predictive Modeling, vol. 26, p. 13. Springer, New York (2013) 15. Scikit Learn Homepage https://scikit-learn.org/stable/index.html. Accessed 21 June 2023

Importance of Blackbody in Everyday Infrared Thermography Hrvoje Glavaš(B) Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Kneza Trpimira 2b, 31000 Osijek, Croatia [email protected]

Abstract. Infrared thermography is a method of detecting thermal radiation, which is then associated with corresponding temperature values. Infrared thermal infrared cameras, like other measuring devices, must undergo a factory and periodic calibration process. The calibration is performed using blackbodies. This paper analyzes the practical application of the blackbody in the calibration of infrared thermal camera. Two infrared thermal cameras Flir E6 and Hikmicro M30 were calibrated with a Voltcraft IRS -350 blackbody. The calibration results are presented in tabular form, while the effect of the blackbody uncertainty on the overall measurement uncertainty is presented in graphical form. Since blackbodies are prohibitively expensive for most thermographers due to their price, calibration guidelines are provided to users when a blackbody is not available and the accuracy of the camera needs to be verified. Keywords: Infrared thermography · Blackbody · Calibration · Temperature measurement · Measurement uncertainty

1 A Short Historical Introduction to Infrared Thermography The human eye can perceive only 36% of the wavelengths emitted by the Sun [1]. The history of infrared thermography begins on April 24, 1800 [2]. On that day, Sir John Herschel (1738 Hanover, Germany - 1822 Slough, England), a professional musician and orchestra conductor [2], published the observation that there is radiation below the visible red part of the spectrum (Latin “infra” - below), which heats up more than visible light radiation. He performed the experiment on February 11, 1800, and the name itself came four decades after the discovery of what Sir John Herschel called dark heat or, more precisely, Calorific Rays. The discovery of the thermoelectric effect in 1821 was a step in the development of thermography. Thomas Johann Seebeck observed an electric current flowing through the junction of two metals when their junctions were at different temperatures. After the discovery of the thermocouple, Macedonio Melloni came up with the idea of forming a series of thermocouples from bismuth and copper. In this way, he increased the sensitivity by 40 times and was able to detect the radiation of a person at a distance of up to 9 m [3]. John Frederick William Herschel (1792. Slough - 1871. Collingwood), son of Sir John and lover of photography, recorded © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 364–374, 2024. https://doi.org/10.1007/978-3-031-51494-4_30

Importance of Blackbody in Everyday Infrared Thermography

365

the first thermogram in 1840 by focusing solar radiation on carbon particles in alcohol with the help of a lens, using the drying process called “evaporography”. In addition to the thermogram, John also demonstrated the existence of different spectral windows of atmospheric transmittance for specific wavelengths [4]. The next major step in the development of thermography was taken by Samuel Pierpont Langley (1834. Roxbury, USA - 1906. Aiken, USA) with the invention of the “Langley bolometer” in 1880, a device that detects electromagnetic radiation by increasing the resistance of conductors. The design was based on the Wheatstone bridge with thin platinum strips [5]. He developed the device for a full 20 years, increasing the sensitivity 400 times over the prototype. Finally, he was able to detect the heat of cattle at a distance of up to 400 m [5]. In 1917, Theodore W. Case performed photon detection using a thallium-sulfur compound (Tl2 S) that changes resistance when exposed to light [6]. Although his work enabled communication over a distance of 29 km through a smoke-filled atmosphere, it did not gain acceptance until 1930 and represents the first thermal photodetection. The development of thermography as we know it today began in the 1940s. It is interesting to note that the study of the infrared part of the spectrum quickly found application in medicine. In 1948, Leo Massopust (1893 - 1970) published a paper on “Infrared photographic examination of the superficial veins of the breast in relation to breast cancer” [7]. The development of the first generation of thermal imaging cameras was shaped by the Pyroscan project, which began in the winter of 1955 at the Kelvin Hughes Research Laboratory (Great Dunmow, Essex) as a mathematical investigation of the possibility of seeing through fog. In the early 1970s, portable systems appeared, but the power supply for the devices was not sufficiently solved. Since 2014, thermography has been widely used thanks to the models launched by the companies Thermal Seek, Therm App and FLIR. In 2016, the first smartphone with a IR camera with a resolution of 80 × 60 pixels appeared, the CAT S60 device.

2 Basic Properties of Thermal Radiation All objects emit electromagnetic radiation, and as the temperature decreases, the wavelength increases, which can be seen in the Fig. 1. When the temperature exceeds 525 °C, we perceive light in addition to the sensation of heat.

Fig. 1. Wavelength range of infrared thermography.

366

H. Glavaš

The characteristic components of the radiation registered by the infrared thermal camera are shown in Fig. 2. If the temperature is higher than that of the environment, the radiation from the object itself is the strongest, followed by the reflected radiation from the environment or from warm objects nearby and the influence of the atmosphere if the camera is far from the object. In the case of calibration, the working distance of the camera is one meter, and the radiation component marked with a red square is not present. The influence of reflected radiation from the environment, if there are no significant sources (for example, elements of the central heating of the building), slowly decreases until temperature of object becomes 150 °C, after which its influence is negligible.

Fig. 2. Radiation registered by an infrared thermal camera.

The most important parameter on which thermal radiation depends is emissivity. Unlike light reflected from a surface, thermal radiation is emitted from the surface of the object, and the shape that directional components produce is represented by the emissivity curve visible in Fig. 3. The recommended imaging angle is shown in green. Imaging perpendicular to the surface may result in reflection by the operator and is not recommended in practice, [8]. Polished metal surfaces will result in significant error in determining the temperature of the object at low shooting angles. In thermographic analysis, it must also be taken into account that the emissivity has different values at individual wavelengths depending on the material, but also different values at different temperatures, as shown in Fig. 4.

3 Infrared Thermal Cameras and Blackbody 3.1 Infrared Thermal Cameras The structure of the infrared thermographic camera can be compared to a digital photo camera, except that instead of a glass lens they have a germanium lens and instead of a CCD sensor they have a microbolometer resistor array. The pixel size is in the range of 15 × 15 µm2 to 55 × 55 µm2 . The basic structure of an infrared thermal imaging camera is shown in Fig. 5, [10].

Importance of Blackbody in Everyday Infrared Thermography

367

Fig. 3. Directional spectral emissivity.

Fig. 4. Wavelength dependence of normal emissivity and temperature dependence of emissivity for different materials, [9].

Fig. 5. Design of infrared thermal camera, [10].

Two infrared thermal cameras, Flir E6 and HikMikro M30 (which are relatively new on the market) were used for this paper. The appearance of the cameras is shown in Fig. 6, and their comparative characteristics are listed in Table 1. Unlike the Flir E6, the Hikmicro M30 has two measurement ranges and is a technical solution that closely resembles the best features of Flir and Fluka, but at a fraction of the price. The M30 has its weaknesses (as does the accompanying software), which is why it is still no better than the competition for scientific research, but it is a good option for field work.

368

H. Glavaš

Fig. 6. Infrared thermal cameras Hikmicro M30 and Flir E6 undergoing calibration.

Table 1. Detailed comparison of infrared thermal cameras used for imaging and analysis. Specifications

Flir E6

HikMicro M30

IR resolution ‘Super IR’

160 × 120

384 × 288 (768 × 576)

NETD

< 60 mK

< 35 mK

FOV

45° × 34°

37.5° × 28.5°

IFOV

5.2 mrad

1.75 mrad

Spectral range

7.5–13 µm

7.5–14 µm

Temp. Range (°C)

− 20 to + 250

− 20 to + 150, 100 to + 550

Accuracy (°C) or (%) of reading

± 2°C or ± 2%

± 2°C or ± 2%

As with all measuring devices, the reliability of the thermal imaging camera can be presented with bathtub-shaped reliability curve, Fig. 7. Daily use, mechanical shocks and temperature stresses can cause damage that can only be detected by regular calibration.

Fig. 7. The bathtub-shaped reliability curve of instruments, [11]

3.2 Black Body Blackbodies are reference sources of radiation. There are various technical solutions that are characterized by different temperature dynamic behavior [12]. Basically, they are heaters with temperature control, a small hysteresis and a precise feedback loop. The

Importance of Blackbody in Everyday Infrared Thermography

369

principle of operation is the same as for a soldering iron, only much more uniform and without significant oscillations. In daily practice, the models shown in Fig. 8 are used. The characteristics of the blackbody used in this work are listed in Table 2 and the device can be seen on the right side of Fig. 7 (black box).

Fig. 8. Different technical versions of the black body. Table 2. Blackbody used for calibration proces. Black body

Voltcraft IRS-350

Temperature range

50 °C to 350 °C

Accuracy

± 0.5 °C at 100 °C, ± 1.2 °C at 350 °C

Stability

± 0.1 °C at 100 °C, ± 0.2 °C at 350 °C

Emissivity of measuring area

0.95

Operating temperature

5 °C to 35 °C

Calibration at the factory is done with several black bodies whose temperature gradually increases. The camera is placed on an arm that automatically transports it from one body to another, as shown on Fig. 9.

Fig. 9. Factory calibration procedure of infrared thermal camera.

By measuring signals corresponding to various temperatures put on technical black bodies, laboratory calibration is carried out. Each pixel’s signal is determined based on

370

H. Glavaš

the blackbody temperature during camera calibration. The calibration curve, as defined by expression (1), is a function that approximates points using constants chosen to best match the Planck’s law function, [13]. T = B/In(R1/(R2 · (S + O)) + F)

(1)

where: T - object temperature in Kelvins S - 16 bit RAW value R1- first constant factor R2- second constant factor B- value range from 1300 to 1600. F- value range from 0.5 to 2. O- offset. These values are a fingerprint of the individual camera and are changed after factory calibration by loading them into the camera’s software.

4 Calibration of Infrared Thermal Camera 4.1 Calibration of Infrared Thermal Camera with Blackbody The aim of calibration is to establish precise quantitative relationships between incident radiation and camera output. A relationship between camera signal and blackbody temperature is thus provided by the calibration procedure. Calibration (in our case) began with a blackbody temperature of 50 °C, which was increased in increments of 10 °C to the maximum value of the blackbody temperature of 350 °C. Figure 10 shows the first thermogram taken with the Flir E6 camera on the left and the out-of-range indication by a yellow triangle with an exclamation point, middle. When the E6 camera registers a signal outside the measurement range, it displays the maximum possible value, next to which there is a red circle with a white x indicating entry into saturation (see Fig. 10 on the right).

Fig. 10. Flir E6 calibration, normal left, out-of-range middle and out-of-range in saturation left.

In contrast to the E6, the M30 camera does not display warning signs, but only a discrete more or less large sign in front of the temperature value. Figure 11 on the right shows the thermogram of a black body set to 150 °C in the temperature Lo range (−20 °C to + 150 °C). The middle set of data shows the thermogram of a black body set to 180 °C in the temperature range (−20 °C to + 150 °C), and since the camera cannot display a

Importance of Blackbody in Everyday Infrared Thermography

371

temperature of 180 °C, it shows > 160 °C. When switching to a higher Hi temperature range (100 °C to + 550 °C), which can be seen in Fig. 11 on the right, temperatures below 90 °C cannot be displayed; in Fig. 11 on the right, the blackbody setting was set to 210 °C.

Fig. 11. M30 calibration, normal termogram on the left, Lo range BB set to 180 °C in the centre and Hi range BB set to 210 °C on the left.

Table 3 shows all registered values during the calibration process and the average deviation for individual cameras for specific temperature ranges. Calibration results can also be represented graphically in the plane of measurement result expectations. In Fig. 12, the expected deviation of the black body ± 0.5 °C (±1.2 °C at 350 °C) and the camera ± 2 °C is marked in gray. The deviation range of the Flir E6 camera is shown in pink and ranges from 0 °C to 3.3 °C. M30 has a deviation from 0.3 °C to 1.1 °C in the lower range and from −0.7 °C to 3.21 °C in the upper range. From the table we can conclude that the deviation of the higher rank is 0.1 °C on average, with the minimum amount in the interval from −0.7 °C to 0.9 °C when the camera is used to capture objects with a higher temperature. 4.2 Calibration of Infrared Thermal Camera with the Reference Objects Calibration checks can also be performed by comparing reference objects with known temperatures and temperatures measured by the thermal camera. Boiling water and melting ice can be used for this purpose. Boiling water has a temperature of about 100 °C when it boils vigorously - a few bubbles rising from the bottom is not enough. Mixing ice with a small amount of water results in a liquid temperature of about 0 °C (ice cubes straight from the freezer are much colder). The emissivity of the camera should be set to 0.96 and the camera should be pointed at the water surface, making sure that condensation does not form on the lens. Figure 13 shows three reference points; two cans, one filled with boiling water and the other with water and ice (left) and a thermogram of the operator (right) (Fig. 13). The measurement in Fig. 13 was made on the surface of the cans, which were painted with black matte paint, and the process of adding boiling water was done in real time, so the temperature measurements are relatively accurate. The third reference point is operator’s body temperature. The temperature of the human body, which was measured in the area of the eye canal, is in average 37 °C, as can be seen from Fig. 13 Camera data are in the specification, operator is healthy and HIKMICRO Analyzer software at moment does not have a thermogram rotation feature.

372

H. Glavaš Table 3. Registered temperature values during the calibration process.

IRS-350

Flir E6

delta E6

M30-Lo

delta Lo

M30-Hi

delta Hi

50

51.5

1.5

51.1

1.1

60

61.7

1.7

60.9

0.9

70

72.4

2.4

71.1

1.1

80

82.5

2.5

81.1

1.1

90

92.2

2.2

90.9

0.9

100

102.9

2.9

100.9

0.9

102.6

2.6

110

113.1

3.1

111.0

1.0

112.4

2.4

120

123.3

3.3

120.8

0.8

122.2

2.2

130

132.9

2.9

130.0

0.0

133.2

3.2

140

143.0

3.0

140.3

0.3

141.0

1.0

150

152.7

2.7

150.5

0.5

151.0

1.0

160

162.6

2.6

> 160

160.5

0.5

170

172.8

2.8

> 160

170.6

0.6

180

182.9

2.9

> 160

180.5

0.5

190

192.9

2.9

> 160

190.4

0.4

200

201.9

1.9

> 160

199.7

-0.3

210

211.4

1.4

> 160

209.4

-0.6

220

221.2

1.2

> 160

220.7

0.7

230

231.1

1.1

230.1

0.1

240

241.2

1.2

240.1

0.1

250

250.0

0.0

249.8

-0.2

260

260.0

260.1

0.1

270

269.6

269.3

-0.7

280

278.6

280.1

0.1

290

280.1

290.1

0.1

300

280.1

299.5

-0.5

310

280.1

310.8

0.8

320

280.1

320.4

0.4

330

280.1

330.9

0.9

340

280.1

339.3

−0.7

350

280.1

350.2

0.2

Average

2.2 °C

0.8 °C

0.6 °C

Importance of Blackbody in Everyday Infrared Thermography

373

Fig. 12. Camera and blackbody measurement uncertainties.

Fig. 13. Three reference points for quick camera calibration checks, water 100 °C, water 0 °C and body temperature 37 °C.

5 Conclusion Infrared thermography as a method of detecting infrared radiation and assigning temperature values to that radiation is a mature and proven technique. There are numerous opportunities for measurement error in determining the actual temperature of the object being examined. The most important is to determine the exact emissivity and to avoid the influence of reflections from the heat sources. The instruments used in this work must be regularly checked for calibration. The presented calibration procedure differs from the procedure in a certified calibration laboratory. For this reason, it is necessary to perform the custom calibration after the initial laboratory calibration as a base point for monitoring the condition of the equipment. The calibration check is most easily performed with a blackbody and is required for companies that use infrared thermal imaging cameras in their daily operations. For individuals who own a camera for personal use, the blackbody represents a major expense that can be avoided by checking in three reference points; for more accurate data, the camera must be sent in for calibration.

References 1. Kreith, F., et al.: Heat and Mass Transfer. Mechanical Engineering Handbook. CRC Press, London (1999) 2. Herschel, W.: Experiments on the refrangibility of the invisible rays of the sun. Philos. Trans. R. Soc. Lond. 90, 284–292 (1800) 3. Scott Barr, E.: The infrared pioneers – II. Macedonio Melloni. Infrared Phys. 2, 67–73 (1962) 4. Dummer, G.W.A., Robertson, J.M.: Medical Electronic Laboratory Equipment 1967–68. Pergamon Press, Oxford (1967) 5. Ring, E.F.J.: The historical development of thermometry and thermal imaging in medicine. J. Med. Eng. Technol. 30(4), 192–198 (2006)

374

H. Glavaš

6. Lovell, D.J.: Cashman Thallous Sulfide cell. Appl. Opt. 10(5), 1003–1008 (1971) 7. Massopust, L.C.: Infrared Photographic study of the superficial veins of the thorax in relation to breast tumours. Surg., Gynaecol. Obstet. 86, 54–58 (1948) 8. Bagavac, P., Krstulovi´c-Opara, L., Domazet, Ž: Pulse phase thermography: impact damage retrieval. Mater. Today: Proc. 5(13), 26578–26583 (2018) 9. Vollmer, M., Möllmann, K.P.: Infrared Thermal Imaging: Fundamentals, Research and Applications. John Wiley & Sons, New Jersey (2018). ISBN: 978-3-527-41351-5 10. Budzier, H., Gerlach, G.: Applications of Thermal Infrared Sensors. Dresden University of Technology. John Wiley & Sons, Ltd., New Jersey (2011). ISBN: 9780470871928 11. Halit, E.: Electronic Portable Instruments: Design and Applications. CRC Press, London (2004). ISBN 0-8493-1998-6 12. Pavkovi´c, D., Lisjak, D., Kolar, D., Cipek, M.: PI/PID Controller relay experiment auto-tuning with extended Kalman filter and second-order generalized integrator as parameter estimators. Tehniˇcki vjesnik 30 (3), 715–723 (2023) 13. Minkina, W., Dudzik S.: Infrared Thermography, Errors and Uncertainties. John Wiley & Sons, Ltd., Hoboken, New Jersey (2009). ISBN: 978–0–470–74718–6

The 5S Method and Its Strategic Determinants Within the Organization of Production Plants Dominika Crnjac Mili´c(B) Faculty of Electrical Engineering, Computer Science and Information Technology in Osijek, Josip Juraj Strossmayer University of Osijek, Kneza Trpimira 2B, 31000 Osijek, Croatia [email protected]

Abstract. Racing for the global advantage, companies have been paying increasing attention to workplace design, employee safety, workplace satisfaction, labor productivity and product quality. One of the methods that can help them achieve these goals is the 5S method, which is also considered a management tool for better organization and management of the workplace. Although the 5S method is applicable to a wide range of workplaces, using published research results, this paper provides a comprehensive overview of its challenges and benefits in manufacturing companies. They are analyzed and synthesized in a new way and special attention is paid to the method steps and its implementation. The paper points to a wide scope of the method application as companies have to meet various standards and present certificates that guarantee the safety and quality of manufactured products. Since managers are constantly trying to find ways to obtain and maintain them, one of the objectives of this paper is to provide an overview and critical evaluation of existing published research results that show the impact of the method application on increasing the product quality, the production process, health and safety. The methods from the qualitative methodological framework were used and research results that contribute to the scientific discourse related to the results that can be achieved by applying the 5S method in the organization of production plants were presented. Keywords: 5S method · production plant · strategic determinants · organization

1 Introduction The 5S method is defined as a management tool, but it also represents the basic principles of workplace organization. According to [1] and [2], it was developed in Japan in the late 1960s and early 1970s. Its implementation is related to the Deming cycle, which includes four key phases: plan, do, check and act. As [3] state, the implementation of the 5S method always starts with the analysis of the existing situation, followed by the identification of the problem, and then targeted procedures are implemented. Initially, the goal of its application was to maintain order in the workplace, but later it resulted in other outcomes, such as the safety of work and the product created by this work, greater satisfaction of employees at work and their greater dedication to work. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 375–385, 2024. https://doi.org/10.1007/978-3-031-51494-4_31

376

D. Crnjac Mili´c

This method is an important segment of the transformation of the management system towards “Lean” management, which [4] and [5] describe as a systematic approach to the organization of the workspace and the workplace with the aim of bringing the efficiency of employees to the highest point. In this way, it is possible to create the expected value for customers while minimizing losses. This is done by using different tools and promoting a different mindset, creating additional value for the employee and the entire company. As [6] stated, in addition to achieving work efficiency, losses and complexity of the work process are optimized. According to [7], for continuous progress related to the change of work methods and improvement of working conditions in production companies, it is necessary to follow a certain framework and management method that allows systematic and easier control of the performance of production processes. Using a qualitative methodological framework, the paper provides an overview of the facts published in a number of scientific sources related to the term 5S method, the way of its implementation, the purpose of its application, the benefits of its application and the results that can be achieved with it. Knowledge about 5S method from published sources of the last fifteen years is analyzed and synthesized in order to gain a broader insight into the meaning of this method and its possible results. The second chapter and third chapter present the theoretical knowledge about the 5S method. In the fourth chapter, the benefits and possible results of applying the 5S method in manufacturing companies are shown, an overview and critical evaluation of the success of the implementation of the 5S method in manufacturing companies is given and the current development and direction of this method is shown. The motivation and the idea to write the paper on this subject arose from the realization of the extraordinary benefits of this method and from the realization that until now there has been no summary overview of its strategic determinants and its benefits. Although this method has been used for many years, its application is very current, as pointed out by various researchers in recent years, for example [8–11]. They highlight the 5S method as the most widespread and fruitful tool of “lean” management of production activities, which is a very suitable means to initiate and achieve the process of continuous improvement, with a tendency to expand the steps in its implementation.

2 Strategic Determinants and the Concept of the 5S Method The most important strategic factors for implementing the 5S method are continuous observation, scanning, recording, analysis, collaboration, and removing everything that is not needed from the work environment in the company. It is important to have the support of the management before starting the preparatory steps and the implementation of this method. It will be the key to motivate the employees for the following changes. The name of the methodology was created according to he five key stages of its implementation. According to the source [12], the name of the 5S method was created from five Japanese words and the order of execution of each phase of the method implementation is shown in (Fig. 1). The first part of the process named “Seiri (Sort)” aims to review materials, tools, and equipment and determine what is unnecessary in the work area. The goal of this phase is to simplify the workplace and the work environment in the plant. This part of

The 5S Method and Its Strategic Determinants

377

Seiri (Sort)

Seiton (Set in order)

Seiso (Shine)

Sektesu (Standardize)

Shitsuke (Sustain)

Fig. 1. Phases of implementation of the 5s method

the process is essential for the subsequent elimination of items that do not belong in a particular workplace or are not to be used there for a while. The goal of the second part of the process named “Seiton (Set in order)” is to remove the unimportant and organize the essential so that the employee can more easily find materials, tools, equipment, etc., when they need them. Everything used in a particular workplace must be organized so that it has its own place where it belongs and must be returned to it after use. This especially applies to the tools and equipment required by the workplace. The place of their belonging must be unambiguously marked, so that every user can recognize it. The places to which they belong must be clearly marked and easily visible. The third part of process named “Seiso (Shine)” is primarily aimed at preventing premature wear of the equipment, but also at the health safety of the employees and the products manufactured in that place. As [13] point out, it is important to pay attention to the personal cleanliness of employees in addition to cleaning machines, all parts connected to them, the workplace, the environment and other areas. Through the fourth part of the process called “Sektesu (Standardize)” the schedule and methodology of cleaning and sorting materials is established. This phase is focused on establishing a sorting and cleaning schedule for each hour and day. Finally, there is the fifth phase of the process named “Shitsuke (Sustain)”, which ensures the continuity of the results obtained and the systematic application of the previous phases of the process.

3 Implementation of the 5S Method First, it is necessary to identify the work area or its parts for which the 5S method is to be applied. Then it is important to prepare the implementation of the 5S method in three steps. The five phases of the 5S method are implemented behind them.

378

D. Crnjac Mili´c

3.1 Preparatory Steps for the Implementation of the 5S Method 1. It is necessary to determine a team of employees who will be responsible for the implementation of the 5S method. The criterion for selection is a good knowledge of the way of work, work phases, work activities and needs of individual workplaces where the 5S method will be introduced. 2. After that, it is necessary to determine the initial state of the space we want to set up. The situation at the workplaces and in their surroundings is documented in such a way that it is described and photographed. The date and the exact place of documentation are recorded. Special attention should be paid to problem areas. 3. In addition, problem areas related to contamination, extra supplies, equipment, and unsorted tools out of place will be highlighted. Unmarked areas that could jeopardise employee safety on the job will also be highlighted. 4. Discussions are held with operators, managers, and supervisors about employee needs, the amount of resources that should be in the workplace at any given time, and equipment and tools. It is necessary to educate employees on how to use the 5S method and its value. It is important to make employees and those who lead them aware that the space should be cleared of anything that is no longer needed in the workplace, and to point out to them that something they no longer need is likely to be needed by someone else in the same condition, refurbished or recycled. 3.2 Stages of Application of 5S Tools 1. Sorting is done to help identify what tools, materials, equipment, etc. are available. At this stage, a picture of the situation related to the identification of problematic parts of the work area is obtained. It is determined whether a single item is in its place, and then can follow another phase in which everything that is present in the room (materials, machines, equipment, parts, tools, furniture, etc.) is marked. 2. Everything that is found is marked, including the floor and walls of the work area. It is important to mark in such a way that all employees understand what is involved and can easily identify what is marked. Uniform markings are placed, with detailed data on the type of items and where they belong. According to [14], the standardized colors for labeling are yellow (keep until a certain time specified on the label), green (necessary for work - keep in a suitable place), and red (unnecessary - remove). 3. It is important to identify safety issues and accordingly to them arrange workplaces and equipment in such a way as to minimize these problems. After that, the area next to workplaces, transportation routes, emergency exits, positions within the plant where vehicles are parked, checkpoints, space for raw materials, space for tools, certain flammable and hazardous raw materials, etc. are marked. The colors we use to indicate this may or may not be legally regulated. It is just important to divide the work area with lines in different colors to ensure better visibility. 4. Cleaning of the premises, tools and machines is carried out. Damages are repaired and disturbances related to installations, necessary construction repairs, etc. are eliminated. The goal of this phase is not only to clean, but also to eliminate the cause of the disorder. Everything that does not belong to a workplace or work area is removed. For example, inventories of anything not needed on a work shift are eliminated and

The 5S Method and Its Strategic Determinants

379

placed in storage for easy access when needed. This phase seeks to increase the level of safety and protection of employees by optimizing the risk of accidents and also reducing the possibility of employee illness by removing contamination. 5. In this stage, the changed condition must be documented in relation to the marked objects, the areas where they are located and the marked space. 6. Standardization of the system allows easier regular control of cleanliness maintenance, which is expected after the implementation of the 5S method. According to [15], for this purpose, it is necessary to develop processes and procedures for assigning responsibilities to employees. It is important to establish a schedule and methods for continuous maintenance of cleaning and sorting. It is desirable to involve employees in the development of standards to develop a culture of behavior and responsibility toward the workplace and the resources they use in their work. Standardization can be aided by recording all procedures performed in the previous phases and attempting to incorporate them into daily routines. It is advisable to use visual aids (colors, marked shelves, etc.) to highlight deviations in the previously established arrangement of things as much as possible. The introduction of an official agreement on the implementation of the 5S method, in which expectations and responsibilities are defined prior to implementation, can also help. 7. System maintenance should not be neglected because everything quickly returns to the beginning. The entire organization should think about how to implement the 5S method. According to [16], the implementation of this phase largely depends on the organizational culture and mindset of the company’s employees. Maintenance should be communicated as a shared responsibility to all those involved in the work processes, considering that old habits are difficult to change. The implementation of these and other phases requires the cooperation of all employees, promotes teamwork, order and discipline. As [17] point out, implementing the 5S methodology also requires audits to identify work in progress and sustain improved activities. Therefore, it is useful to introduce checklists that can be conducted on a monthly or weekly basis. It is important to audit the workplaces from time to time, systematically evaluating the elements designed in accordance with the control and preparation and implementation steps. According to [18], setting the timing and number of inspections over a period of time is useful to minimize employees’ resistance to change and help them get used to keeping their workplaces and spaces in order. Dedicated work and persistence are important to ensure that this phase of the 5S method is not missed. According to [19], the development, implementation and continuous monitoring of the implementation of the phases of the 5S method is the responsibility of lean management, not auditors, because it is easier for them to see the improvement or deterioration of the situation.

4 Usefulness and Possible Results of Applying the 5S Method in the Production Plants The application of the 5S method compared to the standard way of working in plants has a number of advantages, such as a more transparent work environment, simplification of workplaces and the work environment. According to [20], clearing the space of non-essential materials, waste, tools and equipment opens the possibility of increasing

380

D. Crnjac Mili´c

production capacity. As indicated by [21] essential equipment and its parts are defined and recorded. The number of unnecessary activities is reduced and, as [22] note, the efficiency, safety, and quality of work are also increased. Everything needed for the work is systematically arranged and placed within easy reach to avoid unnecessary movements and actions as much as possible while employees are working. No time is wasted searching for equipment, materials, and tools, and employee work is maximized. According to [23], due to the increased free space in the facility, faster delivery of goods that go into the product is ensured, and thus the transportation of what is the result of work in a particular workplace. As stated by [24], after applying this method, not only the time saved changes, but also the attitude towards the workplace, encouraging employees to take initiatives to improve work and working conditions in the future. Routines and clear standards are established to keep the workplace in order. According to [25], operational stability is achieved, which is necessary to create and maintain continuous improvements in work. Employees feel more comfortable in their workplace because they have an orderly environment. According to [26], workplace safety is increased by solving the problem of a overcrowded workplace. With tools and materials more readily available, employees also have to be less physically active, resulting in fewer injuries during work. The number of sick days and absences from the workplace in general is reduced. Continuous maintenance of tools and equipment can not only protect employee health, but also prevent unexpected breakdowns that lead to lost production and failure to meet production schedules. As noted in [27], employee morale is boosted by better organization and continuous work control. In [28], the 5S method is described as a philosophy that develops discipline and work standards among employees, and they consider employee attitude as a critical component for the successful implementation of the 5S method to achieve the desired competitiveness. According to [29], management should provide employees with the necessary financial and moral support to perform their tasks so that they have a positive attitude towards the demands placed on them. [30] point out that in production, there is less waste, the quality of products increases, and more efficient and effective work is achieved. The above may imply a more significant satisfaction of product delivery deadlines, given that less time is lost for the implementation of operations in production. Management detects possible problems more easily and can be more agile in removing them or minimizing their consequences. According to [31], it is easier to comply with standards related to product quality, and in addition to protecting the health of employees, the level of health protection of users of manufactured products can be increased. Ultimately, the continuous application of this method affects the competitiveness of companies in the market and higher profitability, as it is easier to harmonize the results of their work with standards such as HACCP, ISO 9001, ISO 45001, ISO 14001, ISO 22000, ISO 50001, ISO 27001, GMP +, SMETA, ECOVADIS, TAPA TSR, FSSC, IFS, BRC, GLOBAL GAP, ASC and others, depending on the type of products manufactured. As [32] stated, they create a positive impression on customers due to the increased level of organization and work efficiency, among other factors.

The 5S Method and Its Strategic Determinants

381

4.1 Prerequisites for Successful Implementation of the 5S Method in Production Facilities Considering the previously analyzed published sources, it can be noted that the 5S method can be successful if: • operators are are well acquainted with the 5S method and recognize its benefits, • the procedure for the identification and elimination of non-essential materials and waste is defined, • concepts of visual signage are used to simplify the management process • the locations for all essential equipment are clearly marked and sufficient space provided for them, • financial resources and the necessary staff training are provided and the system for verifying the implementation of the 5S method has been established, • all employees practice the 5S method regularly and it is self-sustaining, the middle management has a positive attitude towards the implementation of the 5S method and encourages employees and all those involved in the process to be engage in improving the performance of the process, • works continuously to change the organizational culture. Systematic control of the continuous implementation of the 5S method is not easy to establish, but it is necessary because employees often resist change. Without it, companies can easily and quickly return to the comfort zone and the old way of working after trying to implement the 5S method. It is desirable to conduct a continuous selfassessment, as well as revision in the form of an audit. Self-assessment checklists should include checking for the presence of unnecessary items in the workplace, checking the organization of the workplace, supplies, storage and tools, and checking cleanliness. The importance of the 5S audit lies more in the verification of the administrative areas according to the 5S methodology, in order to ensure the organization of all workplaces according to the company standard. The evaluation of the individual elements monitored in relation to the production activity allows for measurability, the detection of deviations, as well as the observation of trends and better comparability with the expected results. To improve the production process, it is also important to map the process with added value, capture problems, analyze the causes and solve problems efficiently. 4.2 Development and Directions of the 5S Method The 5S method tends to expand through additional phases of implementation. Thus, the “Safety-Security” phase was added, i.e. “Safety Monitoring”, through which all areas of the industrial plant are thoroughly inspected, analyzing the risks at each workplace. An extension of the 5S method is called the 6S method. An extension of the 5S method is the 6S method. As mentioned earlier and also pointed out by [33], the phases of the 5S method applied so far have indirectly led to increased workplace safety, but the increasing need for formalized certifications related to the safety of workers and manufactured products has caused the need for an additional phase of the existing method. It enables monitoring the provision of workers with protective equipment according to the requirements of the work tasks they perform. It also guarantees safety at the workplace,

382

D. Crnjac Mili´c

which is in accordance with standards related to the health of employees, safe use of tools and machines without injuries. It aims to expand the scope of the 5S method to meet the needs of occupational safety and health with machinery, which are necessary to increase the efficiency of production processes. This additional phase, added to the already existing one of the 5S method, will make it easier for companies to ensure that their employees continuously use the necessary protective equipment that will adequately protect them from different risks while performing their activities. As [34] noted, it is difficult to achieve similar results in terms of minimizing or avoiding these risks with collective protection measures or work organization measures. In the future, the 5S method will probably be extended to include some steps of the 6S method and beyond, but it is important not to lose sight of the main objective, which is to focus on increasing the company’s profitability and competitiveness in the market. As indicated in [35] its usefulness can be increased by integrating its application with other methods that have already been tested in business practice, Demand Forecasting, BPM - Business Process Management, and Max-mini Policy. As [36] noted, considering the fact that the results of changes in the way of working are often not immediately noticeable to employees and managers, and that the application of each new step of existing methods or the addition of new methods in daily work can be considered as additional work, there is a need for continuous training that can ensure their easier and faster implementation and thus a faster visible result of their application.

5 Conclusion It is pointed out that the implementation of the 5S method can bring various benefits to companies, such as product safety and quality management, information security, safe environment, protection and maintenance of employee health, and more. It can also affect the reduction of the time of the production cycle or the performance of other work activities, so that the time to perform the activities in a particular workplace is reduced. The workspace is cleared of the unnecessary, minimizing the time spent searching for what is needed during the production process. More space is left for things that can be used in addition while performing work activities. This alone provides a quality work environment, a consistent and reviewed workflow. Work safety and ergo-nomics of workplaces are increased. Because the 5S method is applied with a visual mindset and is based on consistency and continuity, employees are able to identify irregularities more quickly and react faster to correct them. Waste is reduced. Standards are established. It promotes self-control among employees and facilitates their involvement in the management process. The application of this method does not involve high costs, but it does require greater employee commitment and a change in thinking about the importance of workplace organization. Work in the plants is made easier because there is no crowd in the space. Less time is spent searching for resources that employees need to do their jobs. Supplies, unnecessary inventory and equipment in the work environment are located and eliminated. Employee movement and transportation through the facility is facilitated. Employees lose less energy on the job, but the possibility of hazards on the job also decreases. This increases employee morale and commitment to the workplace and work processes. In addition to factories, the system can be used in other areas of the business,

The 5S Method and Its Strategic Determinants

383

such as maintenance shops, tool rooms, warehouses, administrative offices, etc. This in itself opens up the possibility for further research that would continuously monitor the published results on the benefits and opportunities to improve implementation and expand through new implementation steps.

References 1. Jaca, C., Viles, E., Paipa-Galeano, L., Santos, J., Mateo, R.: Learning 5S principles from Japanese best practitioners: case studies of five manufacturing companies. Int. J. Prod. Res. 52(15), 4574–4586 (2014) 2. Liker, K.: The Toyota Way Fieldbook. McGraw-Hill Professional, Australia (2006) - c, D.: Proces implementacije lean-a u malim organizacijama. 3. Piškor, M., Kondi´c, V., Maderi´ Tehniˇcki glasnik 5(1), 103–108 (2011). https://hrcak.srce.hr/clanak/127734. Accessed 28 Mar 2023 4. Bertagnolli, F.: Lean Management. Springer, Cham (2018) 5. Hobbs, P.D.: Lean Manufacturing Implementation: A Complete Execution Manual for any Size Manufacturer. J. Ross Publishing, Fort Lauderdale (2004) ˇ 6. Cabarkapa, J., Sekulovi´c, A.: Lean razmišljanje u proizvodnoj organizaciji–pregled Lean alata za unapredenje procesa. Zbornik Medunarodnog kongresa o procesnoj industriji–Procesing 32(1), 359–372 (2019). http://izdanja.smeits.rs/index.php/ptk/article/view/5005. Accessed 20 Mar 2023 7. Badea, F., Burdus, E.: Contributions on the Lean Management in the current evolution of a company. Econ. Mag., Manage. Ser. 12(1), 168–179 (2009). https://www.management.ase. ro/reveconomia/2009-1/13.pdf. Accessed 17 Mar 2023 8. Hardt, F., Kotyrba, M., Volna, E., Jarusek, R.: Innovative approach to preventive maintenance of production equipment based on a modified TPM methodology for industry 4.0. Appl. Sci. 11(15), 6953 (2021). https://www.mdpi.com/2076-3417/11/15/6953. Accessed 17 June 2023 9. Shahriar, M.M., Parvez, M.S., Islam, M.A., Talapatra, S.: Implementation of 5S in a plastic bag manufacturing industry: a case study. Clean. Eng. Technol. 8, 100488 (2022). https:// www.sciencedirect.com/science/article/pii/S2666790822000933. Accessed 12 Apr 2023 10. Ulhaq, I., George, M., Nayak, R.: 5S and its implications in fashion and textile industry. In: Nayak, R. (eds.) Lean Supply Chain Management in Fashion and Textile Industry. Textile Science and Clothing Technology, pp.125–144. Springer, Singapore (2022). https://doi.org/ 10.1007/978-981-19-2108-7_6 11. Zvidzayi, J.: Reducing manufacturing barriers by introducing a 5S hybrid management system in SA industries. In: Proceedings of the International Conference on Industrial Engineering and Operations Management, Rome, Italy, August 2–5, pp. 2037–2046 (2021). http://ieomso ciety.org/proceedings/2021rome/651.pdf. Accessed 12 July 2023 12. Kaushik, P., Khatak, E.N., Kaloniya, J.: Analyzing relevance and performance of 5S methodology: a review. Int. J. Adv. Res. Eng. Appl. Sci. 4(4), 21–33 (2015) 13. Michalska, J., Szewieczek, D.: The 5S methodology as a tool for improving the organization. J. Achievements Mater. Manuf. Eng. 24(2), 211–214 (2007). http://jamme.acmsse.h2.pl/pap ers_vol24_2/24247.pdf. Accessed 12 May 2023 14. Randhawa, J.S., Ahuja, I.S.: 5S implementation methodologies: literature review and directions. Int. J. Prod. Qual. Manage. 20(1), 48–74 (2017). https://doi.org/10.1504/IJPQM.2017. 080692. Accessed 22 Mar 2023 15. Patel, V.C., Thakkar, D.H.: Review on implementation of 5S in various organization. J. Eng. Res. Appl. 4(3), 774–779 (2014)

384

D. Crnjac Mili´c

16. Ghodrati, A., Zulkifli, N.: A review on 5S implementation in industrial and business organizations. IOSR J. Bus. Manage. 5(3), 11–13 (2012). https://citeseerx.ist.psu.edu/docment? repid=rep1&type=pdf&doi=132feb7900942ce8486065c111f5d683d2ae1054. Accessed 01 July 2023 17. Filip, F.C., Marascu-Klein, V.: The 5S lean method as a tool of industrial management performances. IOP Conf. Ser.: Mater. Sci. Eng. 95(1), 012127 (2015). https://doi.org/10.1088/ 1757-899X/95/1/012127/pdf. Accessed 20 Apr 2023 18. Rizkya, I., Syahputri, K., Sari, R.M., Siregar, I.: 5S implementation in welding workshop–a lean tool in waste minimization. IOP Conf. Ser.: Mater. Sci. Eng. 505(1), 012018 (2019). https://doi.org/10.1088/1757-899X/505/1/012018/meta. Accessed 15 Mar 2023 19. Filip F. C.: Theoretical researches regarding management methods and techniques for conducting the internal audit. In: Annals of DAAAM for 2010 & Proceedings of the 21st International DAAAM Symposium 21, pp.1481–1482 (2010) 20. Bharambe, V., Patel, S., Moradiya, P., Acharya, V.: Implementation of 5S in industry: a review. Multi. Int. Res. J. Gujarat Technol. Univ. 2(1), 12–27 (2020). https://www.gtu.ac.in/GTU-Res earchJournals/News/PAPER%20-%202.pdf. Accessed 15 Apr 2023 21. Arslankaya, S. and Atay, H.: Maintenance management and lean manufacturing practices in a firm which produces dairy products. Procedia-Soc. Behav. Sci. 207, 214–224 (2015). https:// www.sciencedirect.com/science/article/pii/S1877042815052234?via%3Dihub. Accessed 05 July 2023 22. Omogbai, O., Salonitis, K.: The Implementation of 5S lean tool using system dynamics approach. Procedia CIRP 60, 380–385 (2017). https://www.sciencedirect.com/science/article/ pii/S2212827117300586. Accessed 12 Mar 2023 23. Wojtynek, L., Kuli´nska, E., Dendera-Gruszka, M., Kuli´nska, K.: Implementation of lean 5S methodology in logistic enterprise. Res. Logist. Prod. 8, 179–187 (2018). https://yadda. icm.edu.pl/baztech/element/bwmeta1.element.baztech-62d9f0c1-b9a7-4bbe-b2cd-6b8228 2e63e0. Accessed 17 May 2023 24. Singh, J., Rastogi, V., Sharma, R.: Implementation of 5s practices: a review. Uncertain Supply Chain Manage. 2(3), 155–162 (2014) 25. 5S / Visual Workplace Handbook Building the Foundation for Continuous Improvement, Product Automation Corporation Information. https://www.gotopac.com/media/pdf/articles/ 5S-Handbook.pdf. Accessed 16 Apr2023 26. Sharma, K.M., Lata, S.: Effectuation of lean tool “5s” on materials and work space effic. Today: Proc. 5(2), 4678–4683 (2018). https://www.sciencedirect.com/science/article/abs/pii/ S2214785317330146. Accessed 15 Apr 2023 27. Ab Rahman, M.N., Khamis, N.K., Zain, R.M., Deros, B.M., Mahmood, W.H.W.: Implementation of 5S practices in the manufacturing companies: a case study. Am. J. Appl. Sci. 7(8), 1182–1189, (2010). https://www.researchgate.net/profile/Wan-Mahmood-2/pub lication/47630867_Implementation_of_5S_Practices_in_the_Manufacturing_Companies_ A_Case_Study/links/54659c1a0cf2052b509f3407/Implementation-of-5S-Practices-in-theManufacturing-Companies-A-Case-Study.pdf. Accessed 11 Mar 2023 28. Kareem, J.A.H., Amin, O.A.Q.H.: Ethical and psychological factors in 5S and total productive maintenance. J. Industr. Eng. Manage. 10(3), 444–475 (2017). http://www.jiem.org/index. php/jiem/article/view/2313/819. Accessed 17 May 2023 29. Roziana, A.N.: 5S Implementation and People Involvement at Muehlbaeur Technologies Sdn. Bhd. Thesis. Universiti Teknikal Malaysia Melaka (2011) 30. Sati, S.A., Adam, A.I.: Evaluating the effectiveness of 5S implementation in the industrial sector. Int. J. Innov. Sci. Res. Tech. 4(10), 804–808 (2019). https://www.researchgate.net/ profile/Abdelmutalab-Adam/publication/336937404_Evaluating_the_effectiveness_of_5S_i mplementation_in_the_industrial_sector/links/5e444d43(21.06.2023.)458515072d96cb41/

The 5S Method and Its Strategic Determinants

31.

32.

33.

34.

35.

36.

385

Evaluating-theeffectiveness-of-5S-implementation-in-the-industrial-sector.pdf. Accessed 17 Mar 2023 Ashraf, S.R.B., Rashid, M.M., Rashid, A.H.: Implementation of 5S methodology in a food & beverage industri: a case study. Int. Res. J. Eng. Tech. 4(3), 1791–1796 (2017). https://www. researchgate.net/profile/Sk-Ashraf-2/publication/315697643_Implementation_of_5S_Met hodology_in_a_Food_Beverage_Industry_A_Case_Study/links/58dcc0be92851c611d430 ae7/Implementation-of-5S-Methodology-in-a-Food-Beverage-Industry-A-Case-Study.pdf. Accessed 15 Mar 2023 Mrugalska, B., Konieczna, M., Wyrwicka, M.K.: How to improve manufacturing process implementing 5S practices: a case study. In: Ahram, T., Karwowski, W., Vergnano, A., Leali, F., Taiar, R. (eds.) Intelligent Human Systems Integration 2020. Advances in Intelligent Systems and Computing, vol. 1131, pp. 1225–1232. Springer, Cham (2020). https://doi.org/10. 1007/978-3-030-39512-4_187 Šk˚urkova, K.L.: Implementation of the 6S method in an industrial enterprise. IOP Conf. Ser.: Mater. Sci. Eng. 1256(1), 012032 (2022). https://doi.org/10.1088/1757-899X/1256/1/ 012032/pdf. Accessed 01 July 2023 Jiménez, M., Romero, L., Fernández, J., Espinosa, M.D.M., Domínguez, M.: Extension of the Lean 5S methodology to 6S with an additional layer to ensure occupational safety and health levels. Sustainability 11(14), 3827 (2019). https://www.mdpi.com/2071-1050/11/14/ 3827. Accessed 17 May 2023 García-Chávez, G.I., del Rosario Carmelo-Mendieta, V., Collao-Díaz, M.F., Quiroz-Flores, J.C.: Inventory management model based on demand forecasting, 5S, BPM and max-min to improve turnover in commercial enterprises. In 2022 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI), pp.1–6 (2022). https://ieeexplore.ieee.org/stamp/stamp. jsp?tp=&arnumber=9953633. Accessed 20 May 2023 Khan, S.A., Kaviani, M.A., J. Galli, B., Ishtiaq, P.: Application of continuous improvement techniques to improve organization performance: a case study. Int. J. Lean Six Sigma 10(2), 542–565 (2019). https://www.emerald.com/insight/content/doi/10.1108/IJLSS-05-20170048/full/pdf?title=application-of-continuous-improvement-techniques-to-improve-organi zation-performance-a-case-study. Accessed 12 May 2023

The Impact of Electric Car Charging on the Power Grid Zvonimir Klai´c(B)

, Filip Ðakovi´c, Mario Primorac , and Krešimir Fekete

Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Kneza Trpimira 2b, 31000 Osijek, Croatia [email protected]

Abstract. In recent years, the number of electric vehicles has been growing rapidly in the European Union, and in the world in general. Electric cars are more efficient than fossil fuel cars, they have better driving characteristics, and most importantly: they do not emit gas while driving. There are still some challenges related to electric cars. Energy storage - batteries - are still under development to increase the range of cars. An even bigger challenge is the car charging infrastructure - charging stations of increasing capacity are being installed to shorten charging times. The growing number of charging stations brings new challenges, and an important question is what impact car charging stations have on the power quality. The paper presents a detailed analysis of the impact of electric car charging stations on the power quality in the distribution power grid. Keywords: Power quality · harmonics · charging station

1 Introduction The level of harmonic load increases significantly with the increased integration of new devices, emphasizing electric vehicles (EV), solar power plants (PV) and energy storages (ES). Mentioned devices generate or require higher harmonic currents in the network and can cause voltage distortion, which presents new challenges for power quality (PQ). [1, 2] PV systems, as the most common distributed generation (DG) systems, can improve the situation in the network, however, apart from the disadvantages related to variable and stochastic generation, one of the biggest disadvantages of PV systems is represented by inverters based on pulse width modulation (PWM). ES and EV chargers also use PWM converters and thus also contribute to the creation of higher harmonics in the radial distributed systems (RDS) [2]. With the rapid growth in the adoption of electric vehicles (EVs), the demand for electric vehicle charging stations (EVCS) has increased significantly. The increasing deployment of electric vehicles (EVs) and the adoption of DC fast charging technology have introduced new challenges for the power quality of low voltage grids [3]. However, the operation of these charging stations can introduce harmonics into the power grid, leading to power quality issues [4, 5]. Harmonic distortions of individual DG, ES and EV charging stations should be below the limit defined by the standard, but in cases where they are connected in RDS, their cumulative effects © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 386–393, 2024. https://doi.org/10.1007/978-3-031-51494-4_32

The Impact of Electric Car Charging on the Power Grid

387

can increase the level of distortion. The standards that define the maximum permitted limits are HRN EN 50160 2012, [6, 7] HRN [8].The European Norm EN 50160:2012 is a European standard that determines the quality parameters of electricity in distribution networks. This standard describes various aspects of electricity such as voltage, frequency, voltage fluctuations and other important parameters that affect the quality of electricity. When it comes to harmonics, the HEN 50160 standard sets limits for harmonics of non-linear loads in the electrical network in order to ensure an acceptable quality of electricity. These parameters are determined in terms of the maximum allowed values of harmonic currents and voltages up to a maximum of 25 harmonics. The specific limits of harmonics depend on the type of network, the voltage of the distribution network and the voltage level. The EN 61000–2-2:2008 standard is part of a series of European standards related to electromagnetic compatibility (EMC) of electrical devices. Specifically, EN 61000– 2-2:2008 focuses on the influence of harmonics on equipment connected to public distribution networks. The specified norm sets guidelines and limit values for harmonics generated by devices connected to the distribution network. This standard applies to equipment with a nominal voltage up to 1000 V for alternating currents up to 60 Hz and defines the limits of permissible values for current and voltage harmonics at different frequencies. It also focuses on classifying devices according to their harmonic characteristics and requires manufacturers to ensure that their devices meet these limits. In addition to harmonics, the standard also covers other aspects of electromagnetic compatibility, such as voltage fluctuations, voltage variations and interharmonics. The main goal is to ensure that devices do not generate excessive electrical interference that could interfere with other devices in the network.

2 Measurement Analysis The measurements were carried out at the two charging stations for electric vehicles shown in Fig. 1, which are located at the rest area, on a Croatian highway. The charging stations are owned by HEP, and they were manufactured by the company EVTRONIC. The first charging station (on the left in the figure) enables the charging of two electric vehicles with a maximum output DC power of 178 kW. It contains two connectors of different standards for compatibility with the connectors of different vehicle models. One of the standard connectors, currently the most popular, is the Combined Charging System type 2 (CCS - Type 2), and the other is the CHAdeMO connector standard. The second charging station enables charging with a DC power of 50 kW CHAdeMO and CCS - Type 2 standards, and with alternating power of 22 kW Type 2 standard. Both charging stations are powered from one distribution cabinet in which measuring equipment for the power quality measurement was installed. The measurement was performed using a network analyzer A. Eberle PQBox-200, Fig. 2. The measurement lasted seven days, and was performed from October 26 to November 2, 2022. The analysis of the measurement results was performed in the A. Eberle WinPQMobil software package and according to EN 50160 norm limitations.

388

Z. Klai´c et al.

Fig. 1. Charging stations for electric vehicles.

Fig. 2. Network analyzer A. Eberle PQBox-200.

The Impact of Electric Car Charging on the Power Grid

389

In Fig. 3, which summarizes the measurement results, it can be seen that all indices of the power quality are in accordance with the requirements of the EN 50160 standard, except harmonics. As the EN 50160 standard includes harmonic voltages up to the 25th order, the EN 61000–2-2 standard was used for the analysis of harmonic voltages above the 25th order. Figure 4 shows the relative values of harmonics (in percentage) with respect to the fundamental voltage frequency.

Fig. 3. Summarized measurement results.

Fig. 4. Relative values of harmonics.

A more detailed analysis of the spectrum of harmonics shows that odd harmonics are dominant and that the measured values of individual voltage harmonics exceed the threshold value indicated by the red line – Fig. 4. Norms take into account the 95 percentile value of the harmonics - red columns in Fig. 4. So, according to norm EN

390

Z. Klai´c et al.

50160, the voltage of the 21st harmonic is too high and according to EN 61000–2-2, the voltages of the 27th, 33rd and 39th harmonics are too high. Tables 1 and 2 show harmonic values that are too high in relation to the limits in EN 50160 and EN 6100–22, respectively. The tables show critical voltage values of harmonics: excessively high values are marked in red, and critical values for 100% of the week are marked in blue (those values exceed the limit values, but the norms do not take them into account). Table 1. Critical values of voltage harmonics according to the EN 50160.

Table 2. Critical values of voltage harmonics according to the EN 61000–2-2.

Harmonic Order 21 22 26 27 29 30 31 33 34 35 39

Limit Valu e 0,30 0,36 0,35 0,20 1,06 0,33 0,97 0,20 0,32 0,83 0,20

Phase L1 95 % 0,92 0,23 0,18 0,46 0,48 0,13 0,38 0,32 0,12 0,36 0,30

Phase L1 100 % 1,25 0,62 0,36 0,51 0,72 0,24 0,83 0,44 0,21 0,54 0,31

Phase L2 95 % 0,34 0,28 0,32 0,31 0,63 0,25 0,53 0,21 0,19 0,41 0,24

Phase L2 100 % 0,69 0,56 0,73 0,41 1,13 0,62 1,43 0,52 0,32 0,87 0,29

Phase L3 95 % 0,53 0,17 0,14 0,32 0,40 0,12 0,30 0,23 0,10 0,22 0,19

Phase L3 100 % 0,92 0,56 0,43 0,65 0,72 0,42 0,64 0,46 0,27 0,39 0,33

All orders of listed harmonics with exceed values are multiples of 3 (21st, 27th, 33rd, and 39th), which is a particular problem. When the load is linear, the three-phase currents are summed to zero in the neutral conductor, while in the case of a non-linear load, the harmonics that are multiples of the number 3 naturally coincide in phase and time, and cumulatively add up in the neutral conductor at a frequency of mostly 150 Hz (the 3rd harmonic is mostly dominant). In addition to overheating of the transformer, it can occasionally cause overheating and burning of the neutral conductor. In the case of non-linear loads, it is usually recommended to use a transformer with a suitable K factor [9] and to use a transformer with a higher output power and to relieve the transformer.

The Impact of Electric Car Charging on the Power Grid

391

Table 2 and Fig. 4 show that the voltages of the 27th harmonic are too high in all three phases, so its values will be analyzed in more detail. Figure 5 presents voltages and currents of the 27th harmonic during 7 days of measurement. Although it is obvious that there is a cause relationship between the currents of the 27th harmonic (lower diagram) and the voltage of harmonics of the same order (upper diagram), it can be seen that the voltage harmonics are too high daily and in periods when there is no current. The limit value for the voltage of the 27th harmonic is 0.2%, and excessively high values were recorded daily from about 6 pm to about 8 am (the next day). It is assumed that these excessively high voltage values of the 27th harmonic are caused by public lighting. The results are similar for the other harmonic orders for which exceeded values were recorded. Therefore, it can be concluded that the charging station for electric cars is not the main cause of excessively high values of voltage harmonics, as evidenced by Fig. 6, which shows the total harmonic distortion of voltage (THDU) and power (P) of the charging station. However, Fig. 5 also shows that the currents of the 27th harmonic cause excessively high voltage values of the same order harmonic. In the measurement week, an average of about two charging sessions per day was recorded. Given the expected growth in the number of electric cars, a greater number of charging sessions can be expected, which will cause even greater problems with harmonics.

Fig. 5. Voltages and currents of the 27th harmonic during 7 days of measurement.

392

Z. Klai´c et al.

Fig. 6. Total harmonic distortion of voltage (THDU) and power (P) of the charging station.

3 Conclusion The growing number of electric cars also places demands on expanding the infrastructure of electric car charging stations. A legitimate question arises: how will the increasing number of different filling stations affect the distribution power network. This paper presents an analysis of the results of measuring the quality of electricity at an electric charging station on a highway in Croatia. The measurement results show that there is an influence on the voltage harmonics in the distribution network - exceeded values of the 21st, 27th, 33rd and 39th harmonics were recorded. The analysis showed that the harmonic currents of the electric charging station cause voltage harmonics of exceeded values, but it was determined that there is an additional primary source of recorded harmonics - probably public lighting, which is an everyday permanent source of harmonics. Charging sessions were relatively rare during the measurement week - about 2 charging sessions per day, so the charging station is currently a secondary source of harmonics. But, by increasing the number of charging sessions (especially during the daylight), the electric car charging station could become the primary source of harmonics.

References 1. Bernet, D., Stefanski, L., Hiller, M.: Integrating voltage-source active filters intogrid-connected power converters—modeling, control, and experimental verification. IEEE Trans. Power Electron. 36(11) (2021). https://doi.org/10.1109/TPEL.2021 2. Borges, C.L.T., Falcão, D.M.: Optimal distributed generation allocation for reliability, losses, and voltage improvement. Int. J. Electr. Power Energy Syst. 28(6), 413–420 (2006). https:// doi.org/10.1016/j.ijepes.2006.02.003 3. Giri, M., Ronnberg, S.K., Bollen, M.H.J.: Observed harmonic levels on the low voltage grid during EV DC fast charging. In: Proceedings of International Conference on Harmonics and Quality of Power, ICHQP, vol. 2022-May (2022). https://doi.org/10.1109/ICHQP53011.2022. 9808798

The Impact of Electric Car Charging on the Power Grid

393

4. Zhang, Y., Li, T., Wang, S., Jiang, L., Han, W., Diao, X.: Safety assessment of charging stations connected to the power grid considering distribution network constraints. In: 2020 IEEE 4th Conference on Energy Internet and Energy System Integration: Connecting the Grids Towards a Low-Carbon High-Efficiency Energy System, EI2 2020, pp. 2852–2857 (2020). https://doi. org/10.1109/EI250167.2020.9347268 5. Feng, C., Li, R., Liu, G.: Research on harmonic analysis and harmonic suppression measures of electric vehicle charging station. In: Proceedings of 2019 IEEE 2nd International Conference on Automation, Electronics and Electrical Engineering, AUTEEE 2019, pp. 71–75 (2019). https://doi.org/10.1109/AUTEEE48671.2019.9033199 6. “EN 50160 - Voltage characteristics of electricity supplied by public,” CENELEC (2007) 7. Voltage characteristics of electricity supplied by public electricity networks (EN 50160:2010) (2012) 8. HRN EN 61000–2–2:2008/A2:2019 / Hrvatski normativni dokument / HRN4You Hrvatski zavod za norme. https://repozitorij.hzn.hr/norm/HRN+EN+61000-2-2%3A2008% 2FA2%3A2019. Accessed 12 July 2023 9. Shah, N.: Harmonics in power systems (2013)

Investigation of the Improvement of the Wheat Endosperm Hardness Assessment Method by Improved Construction of the Grain Cutting Knife Vinko Krstanovi´c , Kristina Habschied , and Krešimir Mastanjevi´c(B) Faculty of Food Technology Osijek, University of J.J. Strossmayer, F. Kuhaˇca 18, 31 000 Osijek, Croatia [email protected]

Abstract. The majority of bread wheat varieties in Croatia belong to the group of falsely hard “red, or winter wheat” characterized by the so-called “marbling” of the grain, i.e. the alternation of glassy and floury zones in the endosperm, which significantly complicates the precision of the standard method of determining the share of glassy grains according to Pohl, ICC method standard 129. Recent research has shown that by improving the construction of the knife (by changing the angle and cutting intensity (the so-called “new knife”), a much clearer “cleaner” cut is obtained, which makes it possible to determine the degree of glassiness of the grain with greater accuracy, especially for the transitional or marmorated endosperms. Guided by this hypothesis, the degree of glassiness of extremely glassy, extremely floury and two transitional (marmorated) types of grains, cut with the “old knife” (standard device) and the “new” knife on the same instrument. The obtained results show that the obtained values for the degree of glassiness do not differ significantly between the old and new knife for extremely glassy and extremely floury wheat, while for marbled grains the measurement values differ significantly between individual evaluators and between different types of endosperm hardness, especially in the area of medium values for grain glassiness (80–20). The conclusion of the research is that when determining the values for grain glassiness of marbling-early hybrids of bread wheat in Croatia, the application of the “new knife” will give significantly more precise and accurate values and facilitate the visual assessment of endosperm glassiness. Keywords: grain glassiness · knife construction · precision

1 Introduction Vitreousness (glassiness) is a physical-chemical and optical property of the endosperm of wheat grains, which, along with hardness, describes the character of its texture. This quality indicator is primarily a varietal (genotypically conditioned) property which is somewhat influenced by agrotechnical measures and weather conditions during the vegetation period. Glassy grains are mechanically harder due to the much higher density of © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 394–401, 2024. https://doi.org/10.1007/978-3-031-51494-4_33

Investigation of the Improvement of the Wheat Endosperm

395

the endosperm compared to floury ones, and have a natural transparency (glassiness) as a visual sign of hardness. This higher endosperm density manifests itself as increased hardness (grain resistance to mechanical shearing during milling) [1–4]. According to the hardness, wheat varieties are divided into hard and soft, although in recent times, efforts have been made to determine several classes of hardness/softness, because many European wheat hybrids are actually transitional forms between hard and soft, i.e. wheat that has a starting value for hardness as hard, and in the industrial process, they behave like soft wheat. Based on glassiness, wheat is classified into three main classes: glassy, floury and partially glassy (transitional, marbled). This phenomenon is called false or transient glassiness because it is lost by soaking the wheat in water. Glassiness of wheat is a qualitative property that, unlike hardness and protein content and structure, can be determined visually. Today, most authors consider that variety that has a proportion of permanently glassy grains >85% [5] to be vitreous. Vitreous wheat differs from floury wheat in the appearance of the endosperm of the grain (starchy and opaque); it is considered to be of higher quality than floury due to the higher quality of semolina protein, the semolina’s yellowish color and uniform, coarser granulation [6]. The most widespread method for simple determination of endosperm glassiness is the standard visual method ICC Method Standard 129, which uses Pohl’s corn cutter measurement technique [7, 8]. This measurement method is simple, does not require complex equipment and can be easily applied in different measurement conditions, but it is destructive (grain cutting). The aforementioned recent test confirmed that significant errors occur during measurement due to the unfavorable influence of the knife design in this standard device, and therefore the aim of this work is to determine deviations during measurements by qualified assessors due to the use of a standard and a new design knife in determining the degree of glassiness of endo-sperm for different types of wheat (glass, floury and transitional or marbled).

2 Materials and Methods The experimental part of the work was to determine the degree of glassiness of different types of wheat (classified by hardness) by five different qualified evaluators using a device for cutting grains according to Pohl and with two different knives (a standard knife and a newly constructed knife) (Figs. 1 and 2). The assessors performed visual measurements in duplicate on both knives for all samples, and the results are given as the mean of the duplicates. Four types of wheat (harvested in 2021/2022.) were selected for testing: Golubica (hard), Tika-Taka and Bezostaja (transitional, marbled) and Indira (floury). Determinations of the degree and nature of glassiness, hardness and other initial quality indicators were determined according to the analytical methods shown in Table 1. As values for the glassy texture of the endosperm, over 80% of its surface has to be determined as glassy and 80 or more out of 100 cut grains are glassy.

396

V. Krstanovi´c et al.

Fig. 1. 3D CAD model of commercially available grain-cutter: (a) grain-cutter assembly; (b) cutting-knife assembly [9].

Fig. 2. 3D CAD model of modified grain-cutter: (a) modified grain-cutter assembly; (b) new design of knife/blade [9].

Investigation of the Improvement of the Wheat Endosperm

397

Table 1. Used methods according to the Middle European Brewing Analysis Commission (MEBAK©) [3] and Analytica-European Brewery Convention (EBC©) [10]. unit

MEBAK®

EBC®

Mass 1000 grains (TGW)

g dm

3.4/4.4

Hardness (NIR-HD)

%

3.2/4.2

Total N

% dm

Starch

% dm

Vitreousness

%

4.1.4.5.1.1 4.9.1 4.1.3.5.1

3 Results and Discussion When considering the initial results of the determined quality indicators of the tested wheats (Table 2), an effort was made to select varieties that display typical hardness values for each type in order to ensure that a sufficiently large gradation in the glassiness of the endosperm is obtained. Namely, it has already been said that the glassiness of the endosperm is closely related to its hardness. This is the reason for the high positive correlation especially between NIR-HD, total vitreous and protein content (Table 2) and, on the other hand, the negative correlation with the starch content. As it can be assumed that the proportion of protein and starch are two quantities that formally add up to 100, so as one increases, the other must decrease, and the glassiness of the endo-sperm is closely related to the proportion and structure of protein, then this correlation is clear. In this way, we ensured that we had one typical vitreous variety (Golubica), one typical floury variety (Indira) and two partially vitreous or marbled varieties (Bezostaja and Tika-Taka). After the total vitrification was determined, the height of transient or false vitrification was also determined for all varieties (Table 2). The results clearly show that this type of vitrification makes up the largest share of the total vitrification of grains, that is, apart from the variety Golubica, which somewhat retained permanent vitrification in the two marbled and floury varieties, the vitrification de-creased significantly after soaking the grains in water (Table 3). It should be noted that even the hardest variety did not reach the recommended proportion of glassy grains of >85%. When visually determining the degree of vitrification of the endosperm by the evaluator for the old and new knife, it was assumed that there would certainly be a big difference in the values for the old and new knife, due to the construction of the knives themselves. Namely, the problem with measuring with an old knife lies in the fact that wheat grains differ greatly in size - large, medium, small; hardness - very hard, medium hard, floury, extremely floury (soft wheat varieties); and shape - (oval, elongated, irregular shape). These differences are the result of differences in the grain texture of each variety, but also due to external influences during the growing season (e.g. drought, mold infection, etc.). In order to obtain a grain section suitable for visual inspection or digital image analysis (DIA), it is necessary to obtain a clear and clean cut. The construction of commercial cutters is such that it is generally not well adapted to the purpose, but often,

398

V. Krstanovi´c et al. Table 2. Basic raw material quality indicators used for the production of wheat mash.

No

ID

TGW

Hardness

V Total vitreous

PV Permanent vitreous

TV Transient vitreous

Protein content

Starch

(g)

NIR-HD

(%)

(%)

(%)

(%)

(%)

1

Golubica

39.9

94.1

64

30

70

13.25

68.75

2

Tika-Taka

44.9

77.2

32

0

100

10.2

71.7

3

Bezostaja

43.6

77

58

10

90

12.6

68.4

4

Indira

43.6

48

12

2

98

9.7

72.7

Table 3. Separated values for the degree and nature of vitreousity of wheat (before and after soaking of 24 h) No

ID

Degree of grain vitreous (%) 100

100–80

80–60

60–40

40–20

20–0

0

1

GOLUBICA before soaking

12

52

20

12

4

0

0

2

after soaking

4

26

22

20

20

8

0

3

TIKA TAKA before soaking

18

14

6

26

12

24

0

4

after soaking

0

0

4

26

32

38

0

5

BEZOSTAJA before soaking

18

40

20

8

10

2

2

6

after soaking

2

8

8

30

32

14

6

7

INDIRA before soaking

0

12

14

20

16

24

14

8

after soaking

2

0

8

8

8

52

22

due to some of the aforementioned deformations of the grain, it actually cuts, breaks or crushes instead. The reason for this lies in the fact that it is actually two plates rubbing against each other, whereby the analyst has to use a lot of force (jerk) to get a grain cut. As a result, the following problems arise: 1) the recoil must be very strong when it comes to glassy wheat, whereby a certain number of grains fall out of the wells (grooves in the plate in which the grains are located); a certain number of grains that have irregular shapes, as stated before, are actually broken or crushed, resulting in a destroyed surface from which it is very difficult to visually determine the proportion of glassy or floury surface, which particularly hinders computer digital image analysis (DIA); 2) as a great force is needed to cut (break) the grain, the moment of force often causes the entire device to

Investigation of the Improvement of the Wheat Endosperm

399

rotate, which requires frequent re-fixing of the device for cutting a new sample, which greatly increases the time required for analysis. Table 4. Mean values of glassiness degree between different wheat varieties classified by hardness of the endosperm using the “old” (SN) and the “new” knife (NN). % Endosperm hardness 100

100–80

80–60

60–40

40–20

20–0

0

SN NN SN

NN SN NN SN NN SN NN SN NN SN NN

10

12

48

52

17

20

17

12

6

4

2

0

0

0

2 Tika-Taka 16 (marbled)

18

10

14

8

6

18

26

31

12

16

24

1

0

3 Bezostaja (marbled)

20

18

38

40

14

20

8

8

15

10

1

2

0

2

4 Indira (soft)

0

0

8

12

16

14

31

20

3

16

26

24

16

14

1 Golubica (hard)

Table 5. Values of % of glassy grains using the “old” (SN) and the “new” knife (NN) determined by different evaluators. reviewer

% Vitreous of endosperm (>80) Golubica (vitreous)

Tika-Taka (marbled, transitional)

Bezostaja (marbled, transitional)

Indira (soft)

SN

NN

SN

NN

SN

NN

SN

NN

1

40

41

22

14

40

34

14

12

2

35

38

20

16

32

40

11

10

3

37

32

13

21

19

36

12

8

4

40

45

27

12

34

52

9

8

5

41

50

10

15

22

36

10

6

This is specifically expressed in the so-called marbled grains (grains with alternating hard and soft zones in the texture, which, when in contact with the cutting board, offer very different resistance to shearing, so certain regions of the endosperm of the grain are either torn or crushed, which affects the reliability of the method, as already stated. By construction of the so-called new knife, this problem has largely been eliminated, although the construction of the knife and the entire cutting board can still be improved [9]. From the results shown in (Tables 4 and 5), it is evident that when cutting and counting completely glassy or completely the floury surfaces of the endosperm in all types of varieties get similar values for the old and new knife.

400

V. Krstanovi´c et al.

Large deviations between the old and the new knife when it comes to the glassiness percentages of the endosperm surface (100; 100–80; 80–60; 60–40; 40–20; 20–0; 0), are observed in the range of 80–20%, where they are most pronounced in varieties with marbled endosperm, as assumed. Scattering of values between individual evaluators is most pronounced in marbled varieties. As a conclusion, it can be stated that the biggest deviation in the estimation of the value (%) of endosperm vitrification occurs in marbled varieties in the % vitrification zone between 80% - 20%, while when using an old knife, significantly higher results are obtained due to the “false” marbling of the endosperm surface caused by the aforementioned defects in its construction.

4 Conclusions The aim of the work was to determine how the new design of the knife for cutting grain in the standard device for determining the degree of endosperm vitrification according to Pohl affects the accuracy of the determination for the transitional type of wheat characterized as “hybrid”, “transitional” or “marbled”, which is characterized by alternation of glassy and mealy zones with a high proportion of “transient” or “false glassiness”. The obtained results show that the obtained values for the glassiness percentage of the endosperm surface do not differ significantly between the old and new knife for extremely glassy and extremely mealy wheat, while for marbled grains the measurement values differ significantly between individual assessors and between different types of endosperm hardness, especially in the area of medium values for the glassiness of the endosperm surface (80–20%). As a conclusion of the research, it is suggested that when determining the glassiness of grains of marbled bread wheat hybrids, the use of the “new knife” will give a more accurate value and facilitate the visual assessment of the glassiness of the endosperm compared to the use of the “old knife”.

References 1. Pasha, I., Anjum, M.F., Morris, C.F.: Grain hardness: a major determinant of wheat quality. Food Sci. Technol. Int. 16(6), 511–522 (2010). https://doi.org/10.1177/1082013210379691 2. Dexter, J.E., Marchylo, B.A., MacGregor, A.W., Tkachuk, R.: The structure and protein composition of vitreous, piebald and starchy durum wheat kernels. J. Cereal Sci. 10(1), 19–32 (1989). https://doi.org/10.1016/S0733-5210(89)80031-1 3. MEBAK (Middle European Brewing Analysis Commission), Band I and Band II. Brautechnische Analysenmethoden, 3rd edn. (4.1.3.5.1), pp. 64–66. Selbstverlag der MEBAK: Freising-Weihenstephan, Germany (1997) 4. Sieber, A.N., Würschum, T., Friedrich, C., Longin, H.: Vitreosity, its stability and relationship to protein content in durum wheat. J. Cereal Sci. 61, 71–77 (2015) 5. Baasandorj, T., Ohm, J.B., Simsek, S.: Effect of dark, hard, and vitreous kernel content on protein molecular weight distribution and on milling and bread making quality characteristics for hard spring wheat samples from diverse growing regions. Cereal Chem. 92(6), 570–577 (2015). https://doi.org/10.1094/CCHEM-12-14-0249-R 6. Konopka, I., Tanska, M., Konopka, S.: Differences of Some chemicals and physical properties of winter wheat grain of mealy and vitreous appearance. Cereal Res. Commun. 43(3), 470–480 (2015). https://doi.org/10.1556/CRC.2014.0048

Investigation of the Improvement of the Wheat Endosperm

401

7. International Association for Cereal Science and Technology (ICC), 129 Method for Determination of the Vitreousness of Durum Wheat (1980), Quality Assurance and Safety of Crops & Foods ISSN 1757-8361, ICC – International Association for Cereal Science and Technology, Vienna, Austria 8. Brankovi´c, G., Dodig, D., Zori´c, M., Surlan-Momirovi´c, G., Dragiˇcevi´c, V., Djuri´c, N.: Effects of climatic factors on grain vitreousness stability and heritability in durum wheat. Turk. J. Agric. For. 38(4), 429–440 (2014). https://doi.org/10.3906/tar-1308-51 9. Mastanjevi´c, K., Habschied, K., Dvojkovi´c, K., Karakaši´c, M., Glavaš H.: Vitreosity as a major grain quality indicator - upgrading the grain-cutter method with a new blade. Appl. Sci. 13(4), 2655 (2023). https://doi.org/10.3390/app13042655 10. European Brewery Convention. Analytica, 5th edn. Fachverlag Hans Carl: D-Nürnberg, Germany (1998)

Computer Network Design for Office Building Environment Damir Blaževi´c(B)

, Tomislav Keser , and Matko Mance

Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek, Kneza Trpimira 2B, Osijek, Croatia [email protected]

Abstract. This paper presents an example of designing a computer network for a business building. When creating the project documentation, care was taken to ensure that it complies with current standards and the Rulebook on technical conditions for the electronic communication network of commercial and residential buildings. The technical description explains what structured cabling is, describes the importance of following its rules and standards, and defines the ways of marking all the equipment used in the creation of this computer network. Furthermore, an overview of all equipment with its characteristics is presented in detail. The logical scheme defines the so-called vertical network, i.e., the backbone, that is, it shows how each individual part of the active equipment inside the building is connected, how the communication cabinets are connected between the floors, and how the network switches and routers inside the communication cabinets are connected to each other. For each socket and cable, a unique label is defined, this creates a horizontal office network. This is the main part of structured cabling, where the precise definition of each individual part of the computer network provides the basis for safe and stable network operation, the possibility of easy upgrades in the future, and the elimination of possible difficulties that may arise. Application of the quality of service (QoS) on the network will be considered is the and recommendations given to avoid undesirable congestion during the flow of data by properly planning the QoS rules, such as packet marking and traffic shaping. Keywords: computer network design · quality of service (QoS) · structured cabling

1 Introduction The topic of this paper is designing a computer network for an office building. The paper will give an overview of the features of the electronic communication network and its components. Care will be taken to ensure that the project complies with current standards and the rulebook on technical conditions for the electronic communication network of commercial and residential buildings [1]. In accordance with the purpose of the building, a suitable level of the electronic communication network will be selected, and the technical description will elaborate all the requirements in detail to a level sufficient for the © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. Keser et al. (Eds.): OTO 2023, LNNS 866, pp. 402–418, 2024. https://doi.org/10.1007/978-3-031-51494-4_34

Computer Network Design for Office Building Environment

403

execution of the electronic and communication infrastructure [3]. Example of connection lists will be shown for each laid cable. It will also describe the interconnections inside the building and the connection to the public electronic and communication network. In the same way, network addressing, and device configuration will be designed, and a logical scheme will be created. Furthermore, the graphic part will show the blueprint of the business building, where all the main elements of the installation will be seen. The building for which a computer network needs to be designed is an office building - in Osijek. The building was built in the 18th century, origiin the old core of Tvrda nally intended for military purposes, and is currently converted into a business building. The building was built using the usual, but now old, construction methods, with very thick walls of 0.5–1.5 m, which makes drilling, cabling, and the propagation of wireless network signals difficult; in some places, larger rooms are partitioned into smaller, modern dry construction techniques, which have more favorable characteristics for all the previously mentioned difficulties. Inside the floor of the building there are offices, meeting rooms, a warehouse, a buffet and sanitary facilities. It is necessary to design a network infrastructure for employees. The network will consist of a network cabinet with network equipment (router, network switches, switch panels), IP telephony, wireless network, desktop and laptop computers, and central and network printers. Wherever possible, it is planned to leave room for upgrading the network equipment, and a larger number of network connections than is currently required is planned.

2 Technical Description of the Office Building’s Computer Network The goal is to build a fixed computer network for an office building. The project is adapted to the needs of users within the building and the purpose of the premises. Cabling is performed according to structured cabling criteria. Cables are distributed through wall installation channels with connectors. The number of sockets per office meets user needs and saturation criteria. UTP cable CAT 6 for 1000Base-T LAN network (1000 Mb/s) is used for the horizontal network. UTP cable CAT 7 for 10GBase-T LAN network (10 Gb/s) is used for the vertical network. Active network equipment and patch panels are housed in a locked communications cabinet to prevent tampering, with front glass doors and side and rear metal panels that can be dismantled to allow easier access. 2.1 Structured Cabling Structured cabling is defined by international standards, and in practice it refers to a necessary part of the installation of a business building or space, as important as electrical installation and lighting. In practice, it is most often described as the installation of a computer and telephone network, which represents the narrower part of the definition. It consists of hubs, cables and sockets. Hubs consist of communication cabinets with socalled passive equipment, switch panels, cable guides, shelves, power strips and active equipment, network routers and switches, converters and others [3–5]. The basic principles by which structured cabling is performed are:

404

D. Blaževi´c et al.

• the number of connection points should be sufficient in relation to the number of workplaces • installed equipment and all its components must be standardized and replaceable, regardless of the manufacturer of the equipment • multipurpose application of connections is the essence of structured cabling. • When designing, the following should be kept in mind: • the network must be unique - the computer and telephone network should be unified and implemented through a unique installation • as a rule, the needs for networks grow over time, therefore the planned installation should always be over-dimensioned than the current needs require; subsequent laying of additional installation is pricey and aesthetically disadvantageous • the performed state must be documented in accordance with the standards for labeling sockets, cables, communication cabinets. 2.2 Equipment Marking System Markings of communication cabinets and their termination endpoints follow the recommendations standards for structured cabling but are adapted to the specifics of the space. Below is a detailed description of the marking system. Physical positions are preceded by a “+” sign. The positions of floors, communication distributors and equipment are shown in layout plans. Each of the floors, as well as the associated physical positions of the equipment on each floor, are marked with a corresponding label. Below is Table 1, which shows the floor designations. Table 1. Designation of floors. Floor

Mark

Attic

+02

1st floor

+01

Ground floor

+00

For example, + 01 – indicates the physical location on the first floor. 2.3 Marking of the Network Distributor The hub of the structured cabling installation consists of communication cabinets used for housing active computer network devices and equipment for connecting structured cabling segments. Below is a description of the functions of the distributor and the way of labeling individual parts of the distributor [1]: • +BD – the main distributor of the building (building distributor) – a node that connects the vertical distributions (the first to the second level of cabling) with the horizontal cable distribution. In the distributor, there is also a CPE (Customer Premises Equipment) device that is used for terminating the WAN network (WAN – English Wide Area Network),

Computer Network Design for Office Building Environment

405

• +FD – Floor distributor – a node that connects horizontal cable distributions (third level of cabling) with connection points in classrooms and other offices. Due to the large number of sockets and the configuration of the building, there are several floor splitters in the building, they are connected by verticals V1, V2… Individual positions within the distributor are defined as follows: • +BDy-PPx-z – y indicates the BD distributor number, PP indicates the patch panel, x indicates its serial number, while z indicates the position on the panel, i.e., the port number. Example: • +BD1-PP1-TO01– represents the physical position that, read from right to left, indicates connection 1 on the switch panel 1 (PP1) in the distributor BD (+BD1), • +BD1-PP2-TO29-AP – represents the physical location that, read from right to left, indicates port 29 for the wireless access point (AP) on patch panel 2 (PP2) in the BD distributor (+BD1). 2.4 Examination of the Installed Passive Network After the completion of the physical work, examination of each individual cable is carried out [5, 6]. The testing involves the measurement of all involved parts on the cable route, which includes the socket, the connection of the cable to the socket, the cable itself with its length and bends and other obstacles on the route, the connection to the switch panel and the socket on the switch panel. The part that is not covered by this measurement are the two connecting UTP cables: one from the socket to the user’s network device (for which, as a rule, ready-made connecting cables of fixed lengths no longer than a few meters are used), and the other on the other side of the laid cable when switching on from the connecting panel there is also a very short connecting cable (0.5 m–2 m) in the communication cabinet itself to reach the network switch. This should be taken into account for long routes approaching the specification limit up to a total of 100 m. For this reason, and in order to leave enough space for connection, when measuring on the device, the maximum length of laid cables is limited to 90 m. The measurement is carried out to determine whether the installed cable meets the minimum parameters of interference, crosstalk, and signal suppression. Parameters such as length, delay, resistance are measured. The device that is used for measurement must be sent for calibration once every 12 months. Figure 1 shows an example of the results of one such measurement. The measurement was carried out with a device manufactured by Fluke Networks, model DSX-5000 CableAnalyzer, using software LinkWare for desktop computers.

406

D. Blaževi´c et al.

Fig. 1. Figure 2 example of the results of a single cable measurement.

2.5 Vertical Network, Logical Diagram of a Computer Network Logical diagram of the network equipment inside the main communication cabinet is shown, how the switch panels, network switches, are connected inside, vertical connection with the 2nd floor communication cabinet, FD2, main router and Internet service provider equipment are hierarchically connected to each other (Fig. 2) [7–9].

Computer Network Design for Office Building Environment

407

Fig. 2. Display of the logical connection diagram in the main cabinet BD1.

2.6 Horizontal Network, Listing the Connectors and Cables of the Building’s Computer Network, and Network Equipment A simplified and shortened list of mounted network sockets and laid network cables for the whole building is shown in Table 2. Table 2. List of network sockets and cables. No

End point A Terminal outlet (Full marking)

End point B Room

Patch panel (Short marking)

Room

Cable

Length (m)

Patch panel +BD1-PP1, floor +00 1

+BD1-PP1-TO01

1

1-1-01

17

1-1-W1

62

2

+BD1-PP1-TO02

1

1-1-02

17

1-1-W2

62

47

+BD1-PP1-TO47

14

1-1-47

17

1-1-W47

25

48

+BD1-PP1-TO48

14

1-1-48

17

1-1-W48

25

…

… (continued)

408

D. Blaževi´c et al. Table 2. (continued)

No

End point A Terminal outlet (Full marking)

End point B Room

Patch panel (Short marking)

Room

Cable

Length (m)

Patch panel + FD2-PP7, floor + 02 328

+FD2-PP7-TO328

78

2-7-328

86

2-7-W328

57

329

+FD2-PP7-TO329

79

2-7-329

86

2-7-W329

60

Total installation cable length for building

4044

Computer network equipment and assigned labels are shown in Table 3. Table 3. Labeling of network equipment. Full marking

Short marking

Description

+BDx

BD1

Building distributor

+FDx

FD2

Floor distributor

+BDx+PPx

PP1 - PP7

UTP patch panel

+BDx+SWx

SW1 - SW7

Ethernet switch

+BDx+Rx

R1

Router

APx

AP1 – AP17

Access Point

TOxxx

TOxxx

Terminal output

Px

Pxx

UTP patch cable

Wxxx

Wxxx

UTP installation cable

Figure 3 shows the symplified topological diagram of the network with examples of the network equipment placement and interconnection. Diagram is created with the Cisco Packet tracer program.

3 Quality of Service (QoS) The need to implement network quality of service occurs when network traffic exceeds the maximum bandwidth of the computer network [10, 11]. The growth of different types of available services and the connection of more and more different types of network devices shows the need to manage network traffic, i.e. to manage the part of network traffic that is not critical or time-limited, such as e-mail, Internet browsing, data exchange, to limit these services in favor of other network traffic that is essential for real-time performance, e.g. telephone calls, VoIP (Voice over IP), video calls and video conferences, video transmission in general [12, 13].

Computer Network Design for Office Building Environment

409

Fig. 3. Network topology.

As different network services require different conditions and procedures for data flow on the network, the following QoS categories are determined [14, 15]: • • • • • • • • • •

voice traffic call signaling traffic interactive video traffic streaming video traffic best effort data bulk data traffic mission critical traffic IP routing data traffic network management traffic unwanted traffic.

For example, voice traffic requires priority data processing and sufficient bandwidth. The amount of data alone is not large for today’s computer networks, but it must be enabled in real time. A successful voice call service requires all four of the b factors below simultaneously; in case of loss of one of the factors, there will be loss of quality, interruptions or breaking of the connection. The minimum possible requirements for successful voice communication: • • • •

data loss must not exceed 1% delay should be less than 150 ms (in one direction) average delay between packets should be less than 30 ms data transmission speed between 21 and 320 kbs, depending on the selected audio transmission quality and VoIP coding.

Additionally, call signaling traffic requires a bandwidth of 150bps per IP phone. The device. The total satisfactory delay (Fig. 4) is 150–200 ms in one direction, from the sound source, i.e., the first speaker, to the second speaker, i.e. the listener. If the delay is further increased, an excessive delay is obtained where it is obvious that the conversation is no longer taking place as live (in real time) [16, 17]. Video content requirements, as well as voice traffic, require priority data processing, with sufficient (now significantly higher) data bandwidth. The video content that is

410

D. Blaževi´c et al.

Fig. 4. Components of VoIP call data transmission delay.

transmitted can be one-way (eng. Streaming), which is less demanding, and two-way, i.e. conference (eng. Interactive) [10]. Table 4 shows a comparative overview of the requirements for the transmission of these two types of video content transmission. Table 4. Comparative overview of the requirements for the transmission for two types of video content transmission. Request

Interactive video

Streaming video

Packet loss

No more than 1%

No more 5%

Delay in one direction

No more than 150 ms

No more than 4–5 s

Inter-packet delay

No more than 30 ms

There are no special requirements for inter-packet delay

Bandwidth reservation

Bandwidth reservation is determined as the amount required for video transmission increased by 20%

Depending on the format and data bit rate

Streaming video traffic consists of packets of variable size and variable bit rate. Figure 5 shows the video packet size distribution. The initial frame (I frame, Intra-coded picture) represents a full picture frame (the largest data size), and the frames that follow P (Predicted picture) are frames that contain information related to changes compared to the previous frame, therefore, a smaller amount of data, and B frames (Bidirectional predicted picture) which contain information related to changes in relation to the previous and next frame and consequently the smallest amount of data. In the case when the image is static and has few changes, packet size will be smaller. When planning QoS rules, it is important to mark packets as close as possible to the end point (user) (Fig. 6). It is also recommended to use DCSP labels (Layer 3) because they remain the same, they do not change with network technology, that is, they remain the same from source to destination (end to end). Then the next network devices can be relieved of this work and perform the application of higher QoS settings [18].

Computer Network Design for Office Building Environment

411

[CELLRANGE] [CATEGORY NAME] [CELLRANGE] [CATEGORY NAME]

[CELLRANGE] [CATEGORY NAME]

[CELLRANGE]

[CELLRANGE][CATEGORY NAME] [CATEGORY NAME]

Fig. 5. Video packet size distribution.

Fig. 6. Recommended packet marking boundaries.

3.1 Layer 2, and Layer 3 Packet Marking Marking at the L2 layer includes setting the value for the Layer 2 802.1Q/p Class of Service (CoS) field. Figure 7 shows the location of the 802.1Q field.

Fig. 7. Location and content of 802.1Q fields.

412

D. Blaževi´c et al.

Where are the fields Tag Protocol Identifier (TPID), Tag Control Information (TCI). The TPID field is 2 bytes in size, i.e. a 16-bit field with a value of 0x8100, which indicates that it is 802.1Q. Marked box. The TCI field is also 2 bytes in size, i.e., 16 bits and consists of three subfields: Priority Code Point (PCP), Drop Eligible Indicator (DEI), VLAN Identifier (VLAN ID). The PCP field is defined by the IEEE 802.1p standard, 3 bits in size and is used to indicate the CoS priority class as shown in Fig. 7 (Table 5). Table 5. Ethernet 802.1Q CoS fields and their meaning. PCP Value/Priority

Traffic Type

0 (lowest)

Background (BK)

1 (default)

Best effort (BE)

2

Excellent effort (EE)

3

Critical applications (CA)

4

Video with