成功大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	詹京哲 Chan, Jing-Jhe
論文名稱：	多感測器製程監控與參數最佳化 Process Monitoring with Multiple Sensor and Parameter Optimization
指導教授：	李家岩 Lee, Chia-Yen
共同指導教授:	陳瑞彬 Chen, Ray-Bing
學位類別：	碩士 Master
系所名稱：	管理學院 - 數據科學研究所 Institute of Data Science
論文出版年：	2020
畢業學年度：	109
語文別：	英文
論文頁數：	64
中文關鍵詞：	線上診斷、參數調整最佳化、製程監控、基因演算法、貝氏最佳化
外文關鍵詞：	In-line Diagnosis, Parameter Tuning Optimization, Process control, Genetic Algorithm, Bayesian Optimization
相關次數：	點閱：110 下載：9
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

在現代製造業中,預防保養與預測保養確保整體製造產線的可靠度長期維持在可用以及合理的範圍,伴隨著對加工精度的需求大幅上升,設備元件以及機械系統等維護、診斷與保養更是日赴重要。

製程參數的設定值會顯著影響產品的良率，而往往製程伴隨著多重品質特性，各個品質特性間具有高度相關性,而與製程參數間呈現複雜的關係。因此，當製程發生偏移時，產線工程師可能無法精準的掌握製程品質與製程參數間的因果關係,並且沒辦法有效率的根據經驗與領域知識來調整參數。實務上，開發一套智慧製造系統來即時的最佳化製程參數且能同時萃取出製程參數異常資訊是必要的。

本研究提出一個健康保養架構，應用在多感測器上的參數最佳化與參數監控。首先,資料蒐集由感測器監控值取代比較難以取得的製程品質特性，並假設感測器與製程品質有高度相關。透過統計製程管制圖監控製程偏移，一旦有感測器值域超過管制界線，就會啟動結合基因演算法(Genetic Algorithm)與貝氏最佳化(Bayesian Optimization)的製程參數調整架構。實驗結果顯示提出的演算法將有效的進行參數調整並增進製程品質與效力。

In the modern manufacturing industry, preventive and predictive maintenance retain the reliability of the whole production system at a functional and reasonable level. As the demand for processing accuracy rising up, the diagnosis and maintenance of equipment components and mechanical system become more and more crucial.

Process quality is significantly affected by the value of manufacturing process parameters (i.e. recipe) which are often having multiple response with highly correlation and show complicated interactions among the process parameters. Therefore, while the manufacturing process shifts, engineers may not catch the causal relationship well and cannot conduct parameters tuning efficiently based on domain knowledge and experience. Developing an advanced intelligent manufacturing system framework to optimize parameters in real-time manufacturing process as well as extract the anomaly information of the process parameters is urgent in practice.

This study proposes a health maintenance framework for parameters optimization and monitoring in multiple process sensors. First, We start with data generating process considering the nature of sensor correlations to address limited collection of the process quality response. We focus on the monitoring of time series sensor data using statistical control chart for anomaly detection. While the sensor value shows out-of-control, the proposed parameter tuning mechanism combining Genetic Algorithm and Bayesian Optimization is triggered to correct the manufacturing process. The experimental results show that the proposed framework can provide effective parameter adjustment and improve the process quality and efficiency.

摘要 i
Abstract ii
誌謝 iii
Table of Contents iv
List of Tables vi
List of Figures vii
Chapter 1. Introduction 1
1.1. Background and Motivation 1
1.2. Problem Description and Research Scope 2
1.3. Research Overview 4
Chapter 2. Literature Review 5
2.1. Run­to­Run (RTR) Process Control 5
2.1.1. MEWMA Controller 6
2.1.2. Statistical Process Control 8
2.2. Sensor Profile Monitoring 8
2.3. Machine Learning for the Optimization of Process Parameters 10
Chapter 3. Methodology 13
3.1. Research Framework 13
3.2. Stable Process Data Collection 15
3.2.1. Define Process Parameters 15
3.2.2. Parameter and Sensor Stable Data Collection 15
3.2.3. Construct Sensor Profile 15
3.3. Surrogate Model Construction 17
3.3.1. Design of Experiment 17
3.3.2. Distance Calculation 18
3.3.3. Model Selection and Construction 19
3.4. Process Parameter Selection 20
3.4.1. Permutation Feature Importance 20
3.4.2. Genetic Algorithm Optimization 20
3.4.3. Penalty Parameter Selection 25
3.5. Process Adjustment with Bayesian Optimization 28
3.5.1. Bayesian Optimization 28
3.5.2. Surrogate Model 28
3.5.3. Acquisition Functions 29
3.6. In­line Control 31
3.6.1. Cumulative Sum (CUSUM) Control Chart 31
3.6.2. Health Maintenance System 31
Chapter 4. Experimental Study 33
4.1. Data Simulation 33
4.2. Model Construction 36
4.2.1. Surrogate Model Selection 36
4.3. Inline Control 36
4.3.1. CUSUM Control Chart with Stable Process 36
4.3.2. CUSUM Control Chart with Anomaly Process 37
4.4. Process Parameter Selection 40
4.5. Process Parameter Adjustment 46
4.5.1. Bayesian Optimization Surrogate Model Selection 46
4.5.2. Parameter Adjustment with Parameter Selection 48
4.5.3. Parameter Adjustment without Parameter Selection 50
Chapter 5. Conclusion and Future Work 53
References 55
Appendix A. CUSUM control 59
A.1. Anomaly Simulation 59
A.2. Application of CUSUM Monitoring 61
A.3. Distribution of CUSUM Monitoring 62
Appendix B. Author's Biography 64
                                    

[1] Chamil Abeykoon. A novel soft sensor for realtime monitoring of the die melt temperature profile in polymer extrusion. IEEE Transactions on Industrial Electronics, 61(12):7113–7123, 2014.
[2] Ameen AbuHanna and Peter JF Lucas. Prognostic models in medicine. Methods of information in medicine, 40(01):1–5, 2001.
[3] Satnam Alag, Alice M Agogino, Mahesh Morjaria, et al. A methodology for intelligent sensor measurement, validation, fusion, and fault detection for equipment monitoring and diagnostics. AI EDAM, 15(4):307–320, 2001.
[4] André Altmann, Laura Toloşi, Oliver Sander, and Thomas Lengauer. Permutation importance: a corrected feature importance measure. Bioinformatics, 26(10):1340–1347, 2010.
[5] Angelos Angelopoulos, Emmanouel T Michailidis, Nikolaos Nomikos, Panagiotis Trakadas, Antonis Hatziefremidis, Stamatis Voliotis, and Theodore Zahariadis. Tackling faults in the industry 4.0 era—a survey of machinelearning solutions and key aspects. Sensors, 20(1):109, 2020.
[6] Khaider Bouacha and Asma Terrab. Hard turning behavior improvement using nsgaii and psonn hybrid model. The International Journal of Advanced Manufacturing Technology, 86(912):3527–3546, 2016.
[7] Eric Brochu, Vlad M Cora, and Nando De Freitas. A tutorial on bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. arXiv preprint arXiv:1012.2599, 2010.
[8] ChiHang Chen. Apply intelligent agent in manufacturing parameter optimization and empirical studycases study of precision forming and ic packaging process. Master’s Thesis of National Tsing Hua University, Department of Industrial Engineering and Engineering Management, 2018.
[9] WenChin Chen and Denni Kurniawan. Process parameters optimization for multiple quality characteristics in plastic injection molding using taguchi method, bpnn, ga, and hybrid psoga. International journal of precision engineering and manufacturing, 15(8):1583–1593, 2014.
[10] In S Chin, Kwang S Lee, and Jay H Lee. A technique for integrated quality control, profile control, and constraint handling for batch processes. Industrial & engineering chemistry research, 39(3):693–705, 2000.
[11] Saneej B Chitralekha and Sirish L Shah. Application of support vector regression for developing soft sensors for nonlinear processes. The Canadian Journal of Chemical Engineering, 88(5):696–709, 2010.
[12] Ronald B Crosier. Multivariate generalizations of cumulative sum qualitycontrol schemes. Technometrics, 30(3):291–303, 1988.
[13] Peter I Frazier. A tutorial on bayesian optimization. arXiv preprint arXiv:1807.02811, 2018.
[14] David E Goldberg, Jon Richardson, et al. Genetic algorithms with sharing for multimodal function optimization. In Genetic algorithms and their applications: Proceedings of the Second International Conference on Genetic Algorithms, pages 41–49. Hillsdale, NJ: Lawrence Erlbaum, 1987.
[15] Harold Hotelling. Multivariate quality control. Techniques of statistical analysis, 1947.
[16] ChinJung Hsu, Vivek Nair, Vincent W Freeh, and Tim Menzies. Lowlevel augmented bayesian optimization for finding the best cloud vm. arXiv preprint arXiv:1712.10081, 2017.
[17] ShaoYen Hung, ChiaYen Lee, and YungLun Lin. Data science for delamination prognosis and online batch learning in semiconductor assembly process. IEEE Transactions on Components, Packaging and Manufacturing Technology, 10(2):314–324, 2019.
[18] YuHsin Hung. Constrained particle swarm optimization and bayesian process monitoring for health maintenance in threemass resonant servo control system with friction. Master’s Thesis of National Cheng Kung University, Institute of Manufacturing Information and Systems, 2019.
[19] Armann Ingolfsson and Emanuel Sachs. Stability and sensitivity of an ewma controller. Journal of Quality Technology, 25(4):271–287, 1993.
[20] Wei Jiang and John V Farr. Integrating spc and epc methods for quality improvement. Quality Technology & Quantitative Management, 4(3):345–363, 2007.
[21] Donald R Jones, Matthias Schonlau, and William J Welch. Efficient global optimization of expensive blackbox functions. Journal of Global optimization, 13(4):455–492, 1998.
[22] Gülser Köksal, İnci Batmaz, and Murat Caner Testik. A review of data mining applications for quality improvement in manufacturing industry. Expert systems with Applications, 38(10):13448–13467, 2011.
[23] HJ KUSHNER. A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise(global approach to parameter space searching in spacecraft tracking systems). 1963.
[24] ChiaYen Lee and ZhaoHong Dong. Hierarchical equipment health index framework. IEEE Transactions on Semiconductor Manufacturing, 32(3):267–276, 2019.
[25] ChiaYen Lee and TsungLun Tsai. Data science framework for variable selection, metrology prediction, and process control in tftlcd manufacturing. Robotics and ComputerIntegrated Manufacturing, 55:76–87, 2019.
[26] ChiaYen Lee, ChaoShian Wu, and YuHsin Hung. Inline predictive monitoring framework. IEEE Transactions on Automation Science and Engineering, 2020.
[27] GilYong Lee, Mincheol Kim, YingJun Quan, MinSik Kim, Thomas Joon Young Kim, HaeSung Yoon, Sangkee Min, DongHyeon Kim, JeongWook Mun, Jin Woo Oh, et al. Machine health management in smart factory: A review. Journal of Mechanical Science and Technology, 32(3):987–1009, 2018.
[28] Jay Lee, Fangji Wu, Wenyu Zhao, Masoud Ghaffari, Linxia Liao, and David Siegel. Prognostics and health management design for rotary machinery systems—reviews, methodology and applications. Mechanical systems and signal processing, 42(1 2):314–334, 2014.
[29] Kai Liu, YangQuan Chen, Tao Zhang, Siyuan Tian, and Xi Zhang. A survey of runto run control for batch processes. ISA transactions, 83:107–125, 2018.
[30] Cynthia A Lowry, William H Woodall, Charles W Champ, and Steven E Rigdon. A multivariate exponentially weighted moving average control chart. Technometrics, 34(1):46–53, 1992.
[31] Christina M Mastrangelo, George C Runger, and Douglas C Montgomery. Statistical process monitoring with principal components. Quality and reliability engineering international, 12(3):203–210, 1996.
[32] Michael D McKay, Richard J Beckman, and William J Conover. A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics, 42(1):55–61, 2000.
[33] Ole J Mengshoel, David E Goldberg, et al. Probabilistic crowding: Deterministic crowding with probabilistic replacement. In Proceedings of the Genetic and Evolutionary Computation Conference, volume 1, pages 409–416. Morgan Kauffman, 1999.
[34] Douglas C Montgomery. Introduction to statistical quality control. John Wiley & Sons, 2007.
[35] Indrajit Mukherjee and Pradip Kumar Ray. A review of optimization techniques in metal cutting processes. Computers & Industrial Engineering, 50(12):15–34, 2006.
[36] Anders Olsson, Göran Sandberg, and Ola Dahlblom. On latin hypercube sampling for structural reliability analysis. Structural safety, 25(1):47–68, 2003.
[37] Ewan S Page. Continuous inspection schemes. Biometrika, 41(1/2):100–115, 1954.
[38] Eunjeong L Park, Jooseoung Park, Jiwon Yang, Sungzoon Cho, YoungHak Lee, and HaeSang Park. Data based segmentation and summarization for sensor data in semiconductor manufacturing. Expert systems with applications, 41(6):2619–2629, 2014.
[39] Alain Pétrowski. A clearing procedure as a niching method for genetic algorithms. In Proceedings of IEEE international conference on evolutionary computation, pages 798–803. IEEE, 1996.
[40] Jian Qin, Ying Liu, and Roger Grosvenor. A categorical framework of manufacturing for industry 4.0 and beyond. Procedia cirp, 52:173–178, 2016.
[41] S Joe Qin and Leo H Chiang. Advances and opportunities in machine learning for process data analytics. Computers & Chemical Engineering, 126:465–473, 2019.
[42] Emanuel Sachs, Albert Hu, and Armann Ingolfsson. Run by run process control: Combining spc and feedback control. IEEE Transactions on Semiconductor Manufacturing, 8(1):26–43, 1995.
[43] Furqan Tahir, Muhammad T Islam, John Mack, John Robertson, and David Lovett. Process monitoring and fault detection on a hotmelt extrusion process using inline raman spectroscopy and a hybrid soft sensor. Computers & Chemical Engineering, 125:400–414, 2019.
[44] ShengTsaing Tseng, RouhJane Chou, and ShuiPin Lee. A study on a multivariate ewma controller. Iie Transactions, 34(6):541–549, 2002.
[45] Gustavo NA Vieira, Martín Olazar, José T Freire, and Fábio B Freire. Realtime monitoring of milk powder moisture content during drying in a spouted bed dryer using a hybrid neural soft sensor. Drying Technology, 37(9):1184–1190, 2019.
[46] Can Wang, Ming Yang, Dianguo Xu, and Hong Wu. A novel integrated identification method of model structure and parameters for drive system. In 2018 IEEE 27th International Symposium on Industrial Electronics (ISIE), pages 101–107. IEEE, 2018.
[47] Dong Wang, KwokLeung Tsui, and Qiang Miao. Prognostics and health management: A review of vibration based bearing and gear health indicators. Ieee Access, 6:665–676, 2017.
[48] Dorina Weichert, Patrick Link, Anke Stoll, Stefan Rüping, Steffen Ihlenfeldt, and Stefan Wrobel. A review of machine learning for the optimization of production processes. The International Journal of Advanced Manufacturing Technology, 104(58):1889–1902, 2019.
[49] William H Woodall and Matoteng M Ncube. Multivariate cusum qualitycontrol procedures. Technometrics, 27(3):285–292, 1985.
[50] Sewall Wright. The roles of mutation, inbreeding, crossbreeding, and selection in evolution, volume 1. na, 1932.
[51] Neeraja J Yadwadkar, Bharath Hariharan, Joseph E Gonzalez, Burton Smith, and Randy H Katz. Selecting the best vm across multiple public clouds: A datadriven performance modeling approach. In Proceedings of the 2017 Symposium on Cloud Computing, pages 452–465, 2017.
[52] Norfadzlan Yusup, Azlan Mohd Zain, and Siti Zaiton Mohd Hashim. Evolutionary techniques in optimizing machining parameters: Review and recent applications (2007– 2011). Expert Systems with Applications, 39(10):9909–9927, 2012.
[53] Ray Y Zhong, Xun Xu, Eberhard Klotz, and Stephen T Newman. Intelligent manufacturing in the context of industry 4.0: a review. Engineering, 3(5):616–630, 2017.
[54] Xundao Zhou, Yun Zhang, Ting Mao, and Huamin Zhou. Monitoring and dynamic control of quality stability for injection molding process. Journal of Materials Processing Technology, 249:358–366, 2017.

校內：2024-01-01公開
校外：2024-01-01公開

簡易檢索 / 詳目顯示

相關論文