MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION FACULTY FOR HIGH QUALITY TRAINING GRADUATION PROJECT MECHANICAL ENGINEERING TECHNOLOGY INTEGRATE ANN-GA INTO IMPROVE PARTICLE CLASSIFICATION ALGORITHM IN AUTOMOTIVE PRODUCTION MOVING FORWARD TO GREEN MANUFACTURING LECTURER: PHAM HUY TUAN STUDENT: TRAN MINH THUAN SKL012639 Ho Chi Minh City, March 2024 MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION FACULTY OF MECHANICAL ENGINEERING GRADUATION PROJECT Project title: “ INTEGRATE ANN-GA INTO IMPROVE PARTICLE CLASSIFICATION ALGORITHM IN AUTOMOTIVE PRODUCTION MOVING FORWARD TO GREEN MANUFACTURING ” Advisor: PHAM HUY TUAN, Assoc. Student name: TRAN MINH THUAN Student ID: 19144205 Class: 19144CL1A Academic year: 2019 - 2023 Ho Chi Minh City, March 2024 HCMC UNIVERSITY OF TECHNOLOGY THE SOCIALIST REPUBLIC OF VIETNAM AND EDUCATION Independence – Freedom– Happiness FACULTY OF MECHANICAL -------- ENGINEERING GRADUATION PROJECT ASSIGNMENT Semester I/ 2023-2024 Advisor: PHAM HUY TUAN, Assoc. Student name: Tran Minh Thuan Student ID: 19144205 1. Project code: CTM-112 - Project Title: “Integrate ANN-GA into improve Particle Classification Algorithm in Automotive Production moving forward to Green Manufacturing” 2.
Initial Materials Provided by The Advisor: Input images and database are based on the doctoral thesis of Dr. Phan Quoc Bao: “Characterization of particles for Improving Cleanability in Automotive Production”. Content of the project: - Developing the ANN-GA algorithm on the database of the PhD thesis: “Characterization of particles for Improving Cleanability in Automotive Production”. - Evaluate the effectiveness of the new algorithm compared to the old classification method.
Final product: - Improved the reliability of the classification algorithm on the platform developed with Taguchi from the input of high-tech manufacturing experts in the laboratory of the Korean car manufacturer Hyundai. - Develop identification software for hardware of metal scrap sorting equipment, applied at the micro size particle, collected from 120 micromet grid filtering to support 20 micromet grid filtering classification - Setting up the basic ANN classification algorithms to move forward into Particle Expert System and Web based Particle Classification System. - Connect ANN and Robotic Automation Classification to form Hardware Set-up 5. Date of assignment: 01/10/2023 6.
Date of submission: 23/03/2024 7. Presentation language: The report: English Vietnamese Presentation protection: English Vietnamese HEAD OF FACULTY HEAD OF MINISTRY ADVISOR (Sign with full name) (Sign with full name) (Sign with full name) Protection is allowed …………………………………………… (Advisor sign with full name) DISCLAIMER Project title: INTEGRATE ANN-GA TO IMPROVE PARTICLE CLASSIFICATION ALGORITHM IN AUTOMOTIVE PRODUCTION MOVING FORWARD TO GREEN MANUFACTURING. - Advisor: PHAM HUY TUAN, Assoc. - Student: TRAN MINH THUAN - Student ID: 19144205 - Class: 19144CL1A - Phone: 0911521739 - Email: 19144205@student.vn - Disclaimer “In this graduate project, I present many examples and proposed modifications derived from texts found in published articles, used to illustrate certain principles.
It's important to note that these examples are chosen to demonstrate specific and seemingly confusing points. Therefore, my edits are intended to clarify some information, while acknowledging that without direct communication with the author of the article, interpretations may differ. Furthermore, I would like to confirm that this graduation project is the product of my personal research and efforts. I do not copy from any published articles without citing the original source.
If there is any violation, I take full responsibility.” Ho Chi Minh City, Ferbruary 19, 2024 Tran Minh Thuan ACKNOWLEDGEMENTS First of all, I would like to express my deep gratitude to my supervisor, Associate Professor Dr. Pham Huy Tuan, who has the professionalism of a doctoral associate professor and the dedication of a teacher. helped me complete the project successfully despite many obstacles, his expertise and insightful advice helped me shape my approach to writing this thesis. Without his persistent help, urging and reminders him, the goal of this project will not be realized.
I would like to express my deepest and most sincere thanks to Dr. Phan Quoc Bao, who allowed me to use my doctoral thesis as the foundation database for my project. Bao not only allowed me to use his research as a basis for integrating Artificial Neural Network and Genetic Algorithm (ANN-GA), but also gave me a great opportunity to gain practical experience and learned valuable lessons while working at his company. His mentorship, leadership and interest and help in my development were a source of inspiration and motivation.
Bao's generosity in sharing his expertise as well as his role as both a teacher and a respected leader have been instrumental in this research and in the my future career. I would like to express my sincere thanks to the teachers of the Faculty of Mechanical Engineering, Ho Chi Minh City University of Technical and Education for their support and assistance. The environment they have built is very conducive to learning and research, providing me with much-needed knowledge and experience to pursue my academic goals. In addition, I would like to thank the management team at Vietnam Metal Hardware Co., Ltd, training me a lot of practical experience in the field of mechanical engineering.
Practicial guidance in metal fabrication at site and support throughout my challenging time was invaluable. I would also like to thank my colleagues, M.A Pham Thanh Cong and Mr. Nguyen Ngoc Hoang Phi , who were indispensable in my research journey. Their collaboration, insight, and enthusiasm for our common research area have been instrumental in driving our research.
Working alongside such dedicated and talented associates is both inspiring and rewarding. Finally, I am indebted to my family for their love, encouragement, and endless sacrifices both physically and mentally. Without their support, this achievement would not have been possible. Ho Chi Minh City, Ferbruary 19, 2024 Tran Minh Thuan ABSTRACT In pursuit of promoting Green Manufacturing in the automotive industry, this research integrates Artificial Neural Network (ANN) Genetic Algorithm (GA), called ANN-GA, to enhance the optimization of Particle Classification Algorithm (PCA) was previously developed by Taguchi method, then upgraded by Programming C++ and Matlab optimization function.
This innovation comes from the need to solve the problem of automotive machinery small chip lodged and complex particles created in various engine processing processes such as casting, anodizing, machining, High-pressure water jet, anodizing, sandblasting, grinding, polishing, and assembly. These processes will induce the rist of creation of burr, cast and chip,. These particles lodge inside the transmission, engine and crankshaft, and then damage the functions of these components, posing risks to drivers, such as: suddenly increase speed until losing control, mal- function, suddenly stop engine, negative impact to valve control systems called TC21, TC35, TC 29… which mainly working based on the electro-magnetic oil sensors. Based on the partical experiment in research of Dr.
Phan Quoc Bao, and Professor Sung Lim Ko from the high-tech manufacturing laboratory at Hyundai Motor Korea, South Korea, and the Precison Machining Lab, Konkuk University, South Korea, from 2009 - 2015, has developed a particle expert system (PES) that proved the possibility to classify burrs, blanks and chips, by using very basic image processing techniques. An advanced model of PCA, consisting of 12 distinct particle types, was found by using ANN-GA. This new approach does not only trains PCA parameters with higher accuracy but also ensures a stable success rate, thereby improving the reliability of particle source identification processing. Even though the whole project result was statiscal evaluation, it showed up the trend of creating particle from different sources, processes of metal fabrication.
These decisions may support the top leader CEO, Managers to make suitable decision to invest into suitable production technology related to improving Cleanability This integration of ANN-GA into PCA is a forward-thinking approach to solving cleanability challenges in automotive manufacturing. It aligns with the principles of Green Manufacturing by providing a methodologically sound, environmentally friendly solution to improve product quality and safety. Here is the discussion from Professor Dornfield at ISGMA 2013 Sheraton Hawaii Waikiki Green Manufacturing International Conference. Ho Chi Minh City, Ferbruary 19, 2024 Tran Minh Thuan TABLE OF CONTENTS CHAPTER 1: OVERVIEW.
Overview of the research. The need for research. The goal of the research. The object and scope of research.
The object of the research. The scope of the research. Approach and research method. 4 CHAPTER 2: BASIC THEORIES.
Cleanability problems and technology. Advanced cleaning methods. Deburring Processes and Cleanability. 7 CHAPTER 3: CHARACTERIZATION AND RECOGNITION OF PARTICLES FOR IMPROVING CLEANABILITY IN AUTOMOTIVE PRODUCTION.
Mechanism of particles generation. Standard for particle classification. Algorithm of classification. 15 CHAPTER 4: INTEGRATING ANN-GA INTO THE CLASSIFICATION ALGORITHM.
Introduce ANN-GA. Artificial Neural Network (ANN). Graphical representation of ANN model. Result and Discussion.
Use GA from ANN data to optimize parameters. Function sim(x) to use ANN network for optimization. Optimization results using ANN-GA. New PCA base on ANN-GA.
32 CHAPTER 5: TESTING AND EVALUATION. Particle Classification System. Particle Classification System. Result particle classification system.
Evaluate the results of the new algorithm based on ANN-GA. 39 CHAPTER 6: NEW METHOD FOR PARTICLE CLASSIFICATION - BOUNDARY MATRIX. 46 CHAPTER 7: CONCLUSION AND FUTURE SCOPE. 50 LIST OF FIGURES Fig.1 Shows particles extracted from transmission by grid filter .2 Intersection between casting surface and drilling holes of Automotive Transmission at control valve .3 SEM analysis of particles with chemical composition and surface structure.1 Cleanability in the design-to manufacturing cycle and main influences[1] .2 Design of intersection holes considering cleanability[3].3 Chip morphology, formed by drilling & milling.4 Real industrial cleanability problem in automotive .5 Morphology of surface after different deburring processes .6 Burrs at different locations after HPWJ & Brushing .7 Burr at window after HPWJ and brushing .1 Samples with 4 areas: casting surface, machined surfaces (drilling and milling) .2 Stable burrs formed along the edge of a workpiece even after deburring process.
Unstable burr after machining or deburring process. Generation of cast debris and cast surface. Variety of chip shape. Standard of particle classification after image processing to get clear shape/boundary of particles.
Definition of shape parameters in (a) burr, (b) cast, (c) chip and (d) filament of brush. Schematic illustration of Particle Classification Algorithm (PCA)[4]. Basic working mechanism of neural networks. MATLAB's Neural Network Toolbox.
Simple diagram explaining the GA algorithm. Global Optimization Toolbox with GA in Matlab. Graphical representation of ANN model. Network achitecture ANN.
Use the Levenberg-Marquardt algorithm to train the ANN. Validation and test data. ANN training results for group 1 and group 2. Comparison of experimental SR% with ANN(Group 1) and ANN(Group 2) and Taguchi predicted SR%.
Contour Plot of SR% vs L2 and L3. Contour Plot of SR% vs R1 and L3. Contour Plot of SR% vs L2 and R1. How to transform when you need to find the maximum value using the minimum value optimization algorithm.
New PCA base on ANN-GA. Flow diagram of Particle Classification System (PCS). Examples of the process of image processing in each particle. Feature extraction with length (L) and width (W) for burr, cast, chip, and filament of brush.
Programming decision-making structure in Matlab. GUI (Graphical User Interface) of Particle Classification System. Graph of classification for particles inside a transmission at the final values of parameters. Formula to calculate SR%.
Confusion matrix of Group 1. Confusion matrix of Group 2. Simple simulation of the particle boundary matrix. Diagram to extract classification using boundary matrix.
Result LargestObject images. Some results of the particle processed(The PDF file shows more than 1100 particle tests). Results when drawing the grain distribution chart of Eigenvalue with more than 1100 particle. Robot Arm Programming to point out positions.
Particles randomly collected at Huyndai Motor Company, Ulsan, S. 49 LIST OF TABLES Table 3. Classes of particles in PCA[4]. Automatically measured shape parameters (L, W, A) of sample particles[4].
L-8 experiments, with 2 trial test results from group 1, 2[3] .
