臺灣博碩士論文加值系統

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。
針對質量撓性系統(機械系統)，長期的運作使得系統逐漸老化，因而引發機械系統的震動以及失去對加工精度的控制，許多文獻提出基於模型的前饋控制方法試圖抑制震動與補償系統性誤差，然而精確估計這些非線性物理模型中的系統參數是必要的，同時也是為了達到精準的控制以及萃取出系統健康程度的重要特徵。
本研究提出一套健康保養架構，應用於非線性的質量撓性伺服控制系統並考慮摩擦力的影響，進行前饋控制、參數估計與參數監控。首先，本研究提出限制式粒子群最佳化演算法(Constrained Particle Swarm Optimization)進行參數估計，藉由機械系統反共振頻率作為限制並給予彈性，使得演算法在迭代時能避免部分區域最佳解，以及在收斂時能寬鬆限制以提升解的品質，另一方面，在長期的機台運作上，由於系統老化時參數變化相對平緩，本研究提出貝氏製程監控(Bayesian Process Monitoring)進行長期系統參數的監控，在系統參數發生劇變時發出警報。
本研究之實證研究以台灣頂尖電子設備製造公司為例，並於皮帶傳動系統上進行實驗，在理論上，限制式粒子群最佳化能避免區域最佳解，並估計出一致的參數，另一方面，透過模擬研究驗證貝氏製程監控在長期機台運作下能有效監控緩慢變化參數的劇變。在實務上，此線上健康保養架構能提供前饋控制、參數估計以及參數監控，使得機台能保持加工的精度，節省機械元件、生產線停滯以及裝設感測器的成本，同時更能偵測系統參數的劇變並找出對應的機械元件。

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.
Aiming at the mass resonant (mass-spring-damper) system, a long-term operation causes the system aging which ends up with system vibration and affects system control accuracy. A number of model-based feedforward control methods were reported to compensate such effects. However, it is a requisite to identify accurate parameters of the nonlinear physical model in order to achieve the precise speed control as well as extract the health condition of the mechanical system.
This study proposes a health maintenance framework in three-mass resonant servo control system with friction, which conducts feedforward control, parameter estimation and monitoring. First, we propose a constrained particle swarm optimization (CPSO) approaches to estimate the system parameters. With relaxing the soft equality constraints of the anti-resonance frequency in the mechanical system, the CPSO can avoid the local optimum during the solution searching process and release the constraints while converging. Second, in the long-term operation, a Bayesian process monitoring is also proposed for detecting the rapid drift of the slowly aging system parameters.
An empirical study of a leading electronics manufacturing company in Taiwan has been conducted to validate the proposed approach with a designed experiment of a belt drive system. Theoretically, mass resonant system ensemble with friction successfully models the main effects of current variation in the mechanical system. Meanwhile, the CPSO algorithm avoids local optimum comparing to variant PSOs, and it provides the reliability and validity of the parameter estimation. On the other hand, Bayesian process monitoring has been validated by a simulation study to detect the instantly drift of the slowly changing parameters in long-term operation. Practically, the proposed framework not only enhances the control accuracy of processing machines but also helps company to save the cost of mechanical components, production downtime, and additional sensors as well as identify the potential mechanical components failure.

中文摘要 I
Abstract III
Table of Contents V
List of Figures VII
List of Tables IX
Terminology and Notations X
Chapter 1. Introduction 1
1.1 Background and Motivation 1
1.2 Problem Description and Research Scope 5
1.3 Research Overview 7
Chapter 2. Literature Review 8
2.1 Feedback and Feedforward Control 8
2.1.1 Feedback Control 8
2.1.2 Feedforward Control 10
2.2 Mass Resonant System 11
2.3 Friction Models 13
2.4 System Identification and Parameter Estimation 17
2.5 Statistical Process Control 19
Chapter 3. Methodology and Empirical Study 20
3.1 Research Framework 20
3.2 Constrained Particle Swarm Optimization 22
3.2.1 Standard Particle Swarm Optimization 22
3.2.2 Linear Decreasing Inertia Weight Particle Swarm Optimization 23
3.2.3 Comprehensive Learning Particle Swarm Optimization 23
3.2.4 Constrained Particle Swarm Optimization 24
3.3 Analysis of Belt Drive System 30
3.3.1 Introduction of Belt Drive System 30
3.3.2 Design of Experiment 31
3.3.3 System Identification 33
3.3.4 Model Construction 34
3.3.5 Data Segmentation and Outlier Detection 37
3.3.5 PSO Settings 37
3.3.6 CPSO Parameter Estimation and Validation 39
Chapter 4. Bayesian Process Monitoring ＆ Health Maintenance 45
4.1 Introduction 45
4.2 Bayesian Process Monitoring for a Univariate Continuous Variable 47
4.2.1 Estimation 49
4.2.2 Implementing Bayesian Process Control 50
4.2.3 Monitoring Scheme 52
4.3 Bayesian Process Monitoring for Multivariate Continuous Variables 53
4.3.1 Estimation 55
4.3.2 Implementing Bayesian Process Control 56
4.3.3 Monitoring Scheme 57
4.4 Simulation Study 58
4.4.1 Simulation Models 58
4.4.2 Simulation Results 59
4.5 Health Maintenance System 63
Chapter 5. Conclusion and Future Research 65
5.1 Conclusion 65
5.2 Future Research 66
References 67
Appendix 74

Al-Bender, F. Fundamentals of friction modeling. In Proceedings, ASPE Spring Topical Meeting on Control of Precision Systems, MIT 117–122 (2010).
Ang, K. H., Chong, G., Member, S. ＆ Li, Y. PID control system analysis and design. IEEE Control Syst. 26, 32–41 (2006).
Aziz, N. A. A., Alias, M. Y., Mohemmed, A. W. ＆ Aziz, K. A. Particle swarm optimization for constrained and multiobjective problems: a brief review. In International conference on management and artificial intelligence IPEDR 6, 146–150 (2011).
Barbosa, H. J. C., Lemonge, A. C. C. ＆ Bernardino, H. S. A critical review of adaptive penalty techniques in evolutionary computation. In Evolutionary constrained optimization 1–27 (Springer, 2015).
Bennett, S. A brief history of automatic control. IEEE Control Syst. 16, 17–25 (1996).
Bossert, D. E., Morris, S. L., Hallgren, W. F. ＆ Yechout, T. R. Introduction to aircraft flight mechanics: Performance, static stability, dynamic stability, and classical feedback control. (American Institute of Aeronautics and Astronautics, 2003).
Brandstaätter, B. ＆ Baumgartner, U. Particle swarm optimization - Mass-spring system analogon. IEEE Trans. Magn. 38, 997–1000 (2002).
Brosilow, C. ＆ Joseph, B. Techniques of model-based control. (Prentice Hall Professional, 2002).
Cao, M., Wang, K. W., Fujii, Y. ＆ Tobler, W. E. A hybrid neural network approach for the development of friction component dynamic model. J. Dyn. Syst. Meas. Control 126, 144–153 (2004).
Clerc, M. Stagnation analysis in particle swarm optimisation or what happens when nothing happens reminder of classical PSO. Tech. Rep. CSM-460 (2006).
Clerc, M. ＆ Kennedy, J. The particle swarm-explosion, stability, and convergence in a multidimensional complex space. IEEE Trans. Evol. Comput. 6, 58–73 (2002).
Colosimo, B. M. ＆ Del Castillo, E. Bayesian process monitoring, control and optimization. (Chapman and Hall/CRC, 2006).
Dali, A., Bouharchouche, A. ＆ Diaf, S. Parameter identification of photovoltaic cell/module using genetic algorithm (GA) and particle swarm optimization (PSO). In 2015 3rd International Conference on Control, Engineering ＆ Information Technology (CEIT) 1–6 (IEEE, 2015).
Davies, A. Management Guide to Condition Monitoring in Manufacture. (Institution of Engineering and Technology, 1990).
De Wit, C. C., Olsson, H., Astrom, K. J. ＆ Lischinsky, P. A new model for control of systems with friction. IEEE Trans. Automat. Contr. 40, 419–425 (1995).
Deb, K. An efficient constraint handling method for genetic algorithms. Comput. Methods Appl. Mech. Eng. 186, 311–338 (2000).
Eberhart, R. ＆ Kennedy, J. A new optimizer using particle swarm theory. In MHS’95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science 39–43 (IEEE, 1995).
Elsawaf, A. ＆ Vampola, T. Passive suspension system optimization using PSO to enhance ride comfort when crossing different types of speed control profiles. J. Traffic Logist. Eng. 3, 129–135 (2015).
Gottlieb, J. Evolutionary algorithms for constrained optimization problems. Tech. Univ. Clausthal, Dep. Comput. Sci. 1, (2000).
Hamamoto, K., Fukuda, T. ＆ Sugie, T. Iterative feedback tuning of controllers for a two-mass-spring system with friction. Control Eng. Pract. 11, 1061–1068 (2003).
Hashimoto, S., Hara, K., Funato, H. ＆ Kamiyama, K. AR-based identification and control approach in vibration suppression. IEEE Trans. Ind. Appl. 37, 806–811 (2001).
Hong, S. W. ＆ Lee, C. W. Identification of linearised joint structural parameters by combined use of measured and computed frequency responses. Mech. Syst. Signal Process. 5, 267–277 (1991).
Hotelling, H. Multivariate quality control. Techniques of statistical analysis. McGraw-Hill, New York (1947).
Hu, X. ＆ Eberhart, R. Solving constrained nonlinear optimization problems with particle swarm optimization. In Proceedings of the sixth world multiconference on systemics, cybernetics and informatics, 203–206 (2002).
Iwasaki, M., Miwa, M. ＆ Matsui, N. GA-based evolutionary identification algorithm for unknown structured mechatronic systems. IEEE Trans. Ind. Electron. 52, 300–305 (2005).
Jain, K., Alt, F. B. ＆ Grimshaw, S. D. Multivariate quality control-a Bayesian approach. In Annual Quality Congress Transactions-American Society for Quality Control 47, 645 (AMERICAN SOCIETY FOR QUALITY CONTROL, 1993).
Jardine, A. K. S., Lin, D. ＆ Banjevic, D. A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mech. Syst. Signal Process. 20, 1483–1510 (2006).
Johansson, K. E. Field Monitoring of Nc-machines: A System Approach. (Linköping Institute of Technology, 1987).
Kabzifiski, J. Oscillations and friction compensation in two-mass drive system with flexible shaft by command filtered adaptive backstepping. IFAC-PapersOnLine 48, 307–312 (2015).
Kennedy, J. Particle swarm optimization. Encycl. Mach. Learn. 760–766 (2010).
Kohler, M., Forero, L., Vellasco, M., Tanscheit, R. ＆ Pacheco, M. A. PSO+: A nonlinear constraints-handling particle swarm optimization. In 2016 IEEE Congress on Evolutionary Computation (CEC) 2518–2523 (IEEE, 2016).
Kouhei, O., Shibata, M. ＆ Murakami, T. Motion control for advanced mechatronics. IEEE/ASME Trans. Mechatronics 1, 56–67 (1996).
Lee, C.-Y., Huang, T.-S., Liu, M.-K. ＆ Lan, C.-Y. Data science for vibration heteroscedasticity and predictive maintenance of rotary bearings. Energies 12, 801 (2019).
Lee, J. et al. Prognostics and health management design for rotary machinery systems - Reviews, methodology and applications. Mech. Syst. Signal Process. 42, 314–334 (2014).
Levin, R. I. ＆ Lieven, N. A. J. Dynamic finite element model updating using simulated annealing and genetic algorithms. Mech. Syst. Signal Process. 12, 91–120 (1998).
Li, Q., Xu, Q. ＆ Wu, R. Low-frequency vibration suppression control in a two-mass system by using a torque feed-forward and disturbance torque observer. J. Power Electron. 16, 249–258 (2016).
Liang, J. J., Qin, A. K., Suganthan, P. N. ＆ Baskar, S. Comprehensive learning particle swarm optimizer for global optimization of multimodal functions. IEEE Trans. Evol. Comput. 10, 281–295 (2006).
Liu, L., Liu, W. ＆ Cartes, D. A. Particle swarm optimization-based parameter identification applied to permanent magnet synchronous motors. Eng. Appl. Artif. Intell. 21, 1092–1100 (2008).
Liu, Y. F., Li, J., Zhang, Z. M., Hu, X. H. ＆ Zhang, W. J. Experimental comparison of five friction models on the same test-bed of the micro stick-slip motion system. Mech. Sci. 6, 15–28 (2015).
Lowry, C. A., Woodall, W. H., Champ, C. W. ＆ Rigdon, S. E. A multivariate exponentially weighted moving average control chart. Technometrics 34, 46–53 (1992).
Łuczak, D. Mathematical model of multi-mass electric drive system with flexible connection. In 2014 19th International Conference on Methods and Models in Automation and Robotics (MMAR) 590–595 (IEEE, 2014).
Martin, K. F. A review by discussion of condition monitoring and fault diagnosis in machine tools. Int. J. Mach. Tools Manuf. 34, 527–551 (1994).
Martin, K. F., Hoh, S. M. ＆ Williams, J. H. Condition monitoring machine tool drives via health indices. Fault Detect. Superv. Saf. Tech. Process. 1991 571–576 (1992).
Mastrangelo, C. M., Runger, G. C. ＆ Montgomery, D. C. Statistical process monitoring with principal components. Qual. Reliab. Eng. Int. 12, 203–210 (1996).
Michalewicz, Z. ＆ Schoenauer, M. Evolutionary algorithms for constrained parameter optimization problems. Evol. Comput. 4, 1–32 (1996).
Nobari, A. S., Robb, D. A. ＆ Ewins, D. J. A new approach to modal-based structural dynamic model updating and joint identification. Mech. Syst. Signal Process. 9, 85–100 (1995).
Nowopolski, K. ＆ Wicher, B. Parametric identification of electrical drive with complex mechanical structure utilizing Particle Swarm Optimization method. In 2017 19th European Conference on Power Electronics and Applications (EPE’17 ECCE Europe) P-1 (IEEE, 2017).
Orlowska-Kowalska, T. ＆ Szabat, K. Neural-network application for mechanical variables estimation of a two-mass drive system. IEEE Trans. Ind. Electron. 54, 1352–1364 (2007).
Orvosh, D. ＆ Davis, L. Using a genetic algorithm to optimize problems with feasibility constraints. In Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence 548–553 (IEEE, 1994).
Popovic, M. R., Gorinevsky, D. M. ＆ Goldenberg, A. A. Fuzzy logic controller for accurate positioning of direct-drive mechanism using force pulses. In Proceedings of 1995 IEEE International Conference on Robotics and Automation 1, 1166–1171 (IEEE, 1995).
Robinson, J. ＆ Rahmat-Samii, Y. Particle swarm optimization in electromagnetics. IEEE Trans. Antennas Propag. 52, 397–407 (2004).
Sen, M. K., Datta-Gupta, A., Stoffa, P. L., Lake, L. W. ＆ Pope, G. A. Stochastic reservoir modeling using simulated annealing and genetic algorithm. SPE Form. Eval. 10, 49–56 (1995).
Shahgholian, G. Modeling and simulation of a two-mass resonant system with speed controller. Int. J. Inf. Electron. Eng. 3, 448 (2013).
Shahgholian, G., Faiz, J. ＆ Shafaghi, P. Analysis and simulation of speed control for two-mass resonant system. In 2009 Second International Conference on Computer and Electrical Engineering 2, 666–670 (IEEE, 2009).
Shi, Y. ＆ Eberhart, R. A modified particle swarm optimizer. In 1998 IEEE international conference on evolutionary computation proceedings. IEEE world congress on computational intelligence (Cat. No. 98TH8360) 69–73 (IEEE, 1998).
Sturm, G. W., Feltz, C. J. ＆ Yousry, M. A. An empirical Bayes strategy for analysing manufacturing data in real time. Qual. Reliab. Eng. Int. 7, 159–167 (1991).
Szabat, K. ＆ Orlowska-Kowalska, T. Vibration suppression in a two-mass drive system using PI speed controller and additional feedbacks - Comparative study. IEEE Trans. Ind. Electron. 54, 1193–1206 (2007).
Throne, R. D. Frequency domain system identification of one, two, and three degree of freedom systems in an introductory controls class. In Proc. American Society for Engineering Education Annual Conference ＆ Exposition, Paper 493, 2005 (Citeseer, 2005).
Valluru, S. K. ＆ Singh, M. Metaheuristic tuning of linear and nonlinear PID controllers to nonlinear mass spring damper system. Int. J. Appl. Eng. Res. 12, 2320–2328 (2017).
Van Geffen, V. A study of friction models and friction compensation. DCT 118, 24 (2009).
van Kampen, A. H. C., Strom, C. S. ＆ Buydens, L. M. C. Lethalization, penalty and repair functions for constraint handling in the genetic algorithm methodology. Chemom. Intell. Lab. Syst. 34, 55–68 (1996).
Wang, C., Yang, M., Xu, D. ＆ Wu, H. A novel integrated identification method of model structure and parameters for drive system. In 2018 IEEE 27th International Symposium on Industrial Electronics (ISIE) 101–107 (IEEE, 2018).
Wang, J. ＆ Sas, P. A method for identifying parameters of mechanical joints. J. Appl. Mech. 57, 337–342 (1990).
Wenjing, Z. Parameter identification of LuGre friction model in servo system based on improved particle swarm optimization algorithm. In 2007 Chinese Control Conference 135–139 (IEEE, 2007). 
Worasucheep, C. Solving constrained engineering optimization problems by the constrained PSO-DD. In 2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology 1, 5–8 (IEEE, 2008).
Worden, K. et al. Identification of pre-sliding and sliding friction dynamics: Grey box and black-box models. Mech. Syst. Signal Process. 21, 514–534 (2007).
Yang, T. ＆ Lin, C. S. Identifying the stiffness and damping parameters of a linear servomechanism. Mech. Based Des. Struct. Mach. 32, 283–304 (2004).
Yoshida, K. ＆ Takamatsu, H. PSO-based model identification of a full-scale CVT drivetrain. In 2015 IEEE/SICE International Symposium on System Integration (SII) 971–976 (IEEE, 2015).
Zambrano-Bigiarini, M., Clerc, M. ＆ Rojas, R. Standard particle swarm optimisation 2011 at cec-2013: A baseline for future pso improvements. In 2013 IEEE Congress on Evolutionary Computation 2337–2344 (IEEE, 2013).
Zeng, W., Gao, H. ＆ Jing, W. An improved particle swarm optimization. Inf. Technol. J. 13, 2560–2566 (2014).
Zheng, Y. Q. Parameter identification of LuGre friction model for robot joints. In Advanced Materials Research 479, 1084–1090 (Trans Tech Publ, 2012).
Oriental Motor, Servo motor features overview (2018) , accessed 17 May 2019.
https://www.orientalmotor.com/servo-motors/technology/servo-motor-features.html
Ian Wright, An engineer's guide to CNC turning centers (2017) , accessed 17 May 2019. https://www.engineering.com/AdvancedManufacturing/ArticleID/14512/An-Engineers-Guide-to-CNC-Turning-Centers.aspx