臺灣博碩士論文加值系統

智慧型控制系統在仿生物演算法策略中，常應用在龐大、非線性及難以估計
的受控系統，為了設計出符合真實情況的控制器常利用仿生物演算法搜尋，但又
易陷入區域極值而非全區域極值。本論文中將提出增強學習型生物演算法技術應
用於最佳化問題之研究，利用兩種不同演算法的特性，以優缺點形成互補的演化
行為，來解決許多不同的問題，並將其應用在各種不同領域之最佳化問題。
首先，我們提出利用LQR 來解決最佳化控制器的問題，SVM 學習處理估計系
統狀態及誤差問題。將這兩種方法混合設計控制器及應用，結果證明SVM 可以提
昇控制器的效率，亦可兼顧最佳化的效果。
在模糊控制器的設計上導入進化式基因演算法，並探討利用SVM的分類特性，
在進化的世代中，選出最佳的染色體當作學習的樣本，使演化過程不致掉入區域
最佳解，研究結果發現不但縮短進化的世代，亦大幅改善原本控制器的效能。
由於目前工業界使用FUZZY-PID 控制的業者還是很多，所以在FUZZY-PID
控制器的設計上我們成功以簡易型FUZZY-PID 控制器運用粒子演算法(PSO)的學
習，再混合利用進化演算法(EP)的再生程序協助進化並產生新的PID 的控制參數，
從馬達控制設計到高階系統控制設計實驗研究證明如此隨機最優化策略不是局限
在特殊控制設計而已，且它可兼顧最佳化的效果。
雖在混合式生物演算法中我們可以證明它的最佳化的效果，但我們希望再加
入Q-Learning 以增強PSO 的進化學習能力，此論文將證明怎樣與Q-Learning 合作
有效搜尋最佳FUZZY-PID 控制參數，研究結果雖演化世代大幅減少，但亦不影響
最佳化控制器設計。

Intelligent control system requires an optimized algorithm to meet the challenge in
designing and developing a controller to meet real-world situations involving large,
nonlinear, and difficult to estimate control system. In addition, issues on the use of
biomimetic search algorithm are discussed, for it easily goes into a region with extreme
values that may not be the optimal. Thus, this research proposed a study to enhance the
learning of the biological algorithm technology for optimizing system design. Using the
characteristics of two different algorithms and complementing algorithm behavior, not
only can it solve many different problems, but also its application can be applied in
various fields of optimization problems.
First, we propose the use of LQR controller to solve optimization problems in the
controller and then SVM classification system was used to estimate the system status
and error issues. The hybrid of these two methods was used to design the controller and
application. The results confirmed that the use of SVM can improve the efficiency of
the controller and also assess the optimum effect.
In the fuzzy controller design, a hybrid evolutionary genetic algorithm and SVM
classification techniques were applied. During evolution, the best chromosomes were
selected as training instances and avoid those falling under optimal solution. The result
is also substantially reducing the evolution of generations to improve the performance of the original controller.
Currently there are still many industries that utilize FUZZY-PID control. Therefore,
this research on FUZZY-PID controller successfully designed a simple FUZZY-PID
controller using Particle Swarm Optimization (PSO) learning and regeneration process
of evolutionary algorithms (EP) to help evolve and a develop new PID control
parameters. From the motor control design to the high-order control system, an
empirical study showed that stochastic optimization strategy is not just limited to the
specific control design but can also estimate the optimization effect.
The hybrid biological algorithm can produce optimum effect, and the research tried
adding Q-Learning to reinforce the evolution learning of PSO. This thesis explains in
detail how Q-Learning takes part in searching for the best FUZZY-PID control
parameters. Research results showed that evolution generations were substantially
reduced, while not affecting the optimal controller design.

摘 要 iv
ABSTRACT vi
致 謝 viii
Content ix
List of Tables xii
List of Figures xiii
Chapter 1. Introduction 1
1.1 Motivation and background 1
1.2 Review of previous research 2
1.3 Dissertation Organization 7
Chapter 2. Combining Support Vector Machines with Linear Quadratic Regulator
Adaptation for the Online Design of an Automotive Active Suspension System 8
2.1 Support Vector Machines 8
2.2 The Optimal Control Problem 12
2.2.1 The Suspension System Dynamic Model 12
2.2.2 State Feedback Regulator Design 14
2.2.3 Optimal control by SVM 15
2.3 Simulation 16
Chapter 3. Using Fuzzy logic Controller and Evolutionary Genetic Algorithms for
automotive active suspension system 23
3.1 Evolutionary Genetic Algorithm 23
x
3.1.1 Chromosome Encoding 23
3.1.2 Adaptive Genetic Algorithm 25
3.2 Integrating Adaptive Genetic Algorithm with Support Vector Machine
Scheme 29
3.3 Simulation 33
Chapter 4. Numerical simulation for Fuzzy-PID controllers and helping EP reproduction
with PSO hybrid algorithm 40
4.1 Particle Swarm Optimization 40
4.1.1 The speed and accumulation factors of the swarm 41
4.2 Hybridization of PSO and Evolutionary Programming 43
4.3 System description and problem formulation 47
4.4 Simulation studies and results 50
4.4.1 The second-order model 50
4.4.2 The third-order model 53
4.4.3 The fourth-order model 56
4.4.4 The Time-Delay system 59
4.4.5 The automotive active suspension system 64
Chapter 5. A PSO-Based Adaptive Fuzzy PID-Controllers 68
5.1 The Reinforcement Evolutionary Strategy 68
5.1.1 The Reinforcement PSO Learning 68
5.1.2 The PSO-PID Evolutionary Computing 73
xi
5.2 The Control System Structure 76
5.3 Simulation studies 79
5.3.1 The system model 79
Chapter 6. Conclusions and future works 85
6.1 Conclusions 85
6.2 Future works 87
References 89

1. Y. Yajima, Linear Programming Approaches for Multicategory Support Vector Machines. European Journal of Operational Research. 162 (2005), 514–531.
2. JAK Suykens, J. Vandewalle, and B. De. Moor, Optimal Control by Least Squares Support Vector Machines. Neural Networks, 14 (2001)
3. H. Wang, D. Pi, and Y. Sun, Online SVM Regression Algorithm-based Adaptive Inverse Control. Neurocomputing, 70 (2007), 952–959.
4. F. Friedrichs, and C. Igel, Evolutionary Tuning of Multiple SVM Parameters. Neurocomputing, 64 (2005), 107–117.
5. D.H. Shou, and J.H. He, Application of parameter-expanding method to strongly nonlinear oscillators, International Journal of Nonlinear Sciences and Numerical Simulation, 8. 1. (2007), 121-124.
6. Z.M. Odibat, and S. Momani. Application of variational iteration method to nonlinear differential equations of fractional order, International Journal of Nonlinear Sciences and Numerical Simulation, 7. 1. (2006), 27-34.
7. T.H.S. Li, C.L. Kuo, and N.R. Guo, Design of an EP-based fuzzy sliding-mode control for a magnetic ball suspension system, Chaos, Solitons & Fractals, 33. 5. (2007), 1523-1531.
8. F. Manuel, P. Perez, and P.M. Manuel, Chaotic and steady state behaviour of a nonlinear controlled gyro subjected to harmonic disturbances, Chaos, Solitons & Fractals, 33. 2. (2007), 623-641.
9. H.T. Yau, C.L. Kuo, and J.J. Yan. Fuzzy sliding mode control for a class of chaos synchronization with uncertainties, International Journal of Nonlinear Sciences and Numerical Simulation, 7. 3. (2006), 333-338
10. S. J. Cho, and Y. S. Choi, Estimation of ride quality of a passenger car with nonlinear suspension. Int. J. Automotive Technology 8, 1, (2007). 103-109.
11. J. Z. Feng, J. Li, and F. Yu, GA-based PID and fuzzy logic control for active vehicle suspension syetem. Int. J. Automotive Technology 4, 4, (2003). 181-191.
12. J. J. Grefenstette, Optimization of control parameters for genetic algorithms. IEEE Trans. Syst., Man, Cybern., (1986). 122-128.
13. X. Guo, Y. Zhou, and D. Gong, Optimization of fuzzy sets of fuzzy control system based on hierarchical genetic algorithms. Proc. IEEE TENCON’02, (2002). 1463-1466.
14. J. R. Jang, Self-learning fuzzy controllers based on temporal back propagation. IEEE Trans. Neural Networks, 3-5, (1992). 714-123.
15. K. Kropp, and U. G. Baitinger, Optimization of fuzzy logic controller inference rules using a genetic algorithms. EUFIT'93-First Technologies, (1993). 1090-1096.
16. Y.-P. Kuo, and T.-H. S. Li, GA-based fuzzy PI/PD controller for automotive active suspension system. IEEE Trans. Industrial Electronics 46, 6, (1999). 1051-1056.
17. M. Shin, and M. Sunwoo, Optimal period and priority assignment for a networked control system scheduled by a fixed priority scheduling system. Int. J. Automotive Technology 8, 1, (2007). 39-48.
18. V. N. Vapnik, Statistical Learning Theory. Wiley. New York. (1998).
19. V. N. Vapnik, An overview of statistical learning theory. IEEE Trans. Neural Networks, (1999). 88–999.
20. V. N. Vapnik, The Nature of Statistical Learning Theory. Springer. New York. (2000).
21. F. Wang, T. Zhang, L. Yang, and B. Zhuo, Steadystate optimization of an internal combustion engine for hybrid electric vehicles. Int. J. Automotive Technology 8, 3, (2007). 361-373.
22. L. Wu, and H.-L. Chen, Complex stochastic wheelbase preview control and simulation of a semi-active motorcycle suspension based on hierarchical modeling method. Int. J. Automotive Technology 7, 6, (2006). 749-756.
23. I. Youn, and A. Hac, Preview control of active suspension with integral action.
Int. J. Automotive Technology 7, 5, (2006). 547-554.
24. F. Yu, J.-Z. Feng, and J. Li, A fuzzy logic controller design for vehicle ABS with a on-line optimized target wheel slip ratio. Int. J. Automotive Technology 3, 4, (2002). 165-170.
25. A. Chaibakhsh, A. Ghaffari, and S.AliA. Moosavian, A simulated model for a once-through boiler by parameter adjustment based on genetic algorithms, Simulation Modelling Practice and Theory 15 (2007) 1029–1051.
26. J.-S. Chiou, and K.-Y. Wang, Application of a hybrid controller to a mobile robot, Simulation Modelling Practice and Theory 16 (2008) 783–795.
27. M. Clerc, The swarm and the queen: towards a deterministic and adaptive particle swarm optimization, in: Proceedings of the ICEC (1999), Washington, DC, (1999), pp. 1951–1957.
28. K. Hirota, Industrial Application of Fuzzy Technology, Springer-Verlag, Berlin,(1992).
29. H.F. Ho, Y.K. Wong, and A.B. Rad, Robust fuzzy tracking control for robotic manipulators, Simulation Modelling Practice and Theory 15 (2007) 801–816.
30. W. Huang, D. Chen, An efficient heuristic algorithm for rectangle-packing problem, Simulation Modelling Practice and Theory 15 (2007) 1356–1365.
31. Y.-T. Juang, Y.-T. Chang, and C.-P. Huang, Design of fuzzy PID controllers using modified triangular membership functions, Information Sciences 178 (2008)1325–1333.
32. J. Kennedy, and R.C. Eberhart, Particle swarm optimization, Proceedings of IEEE International Conference on Neural Networks IV (1995) 1942–1948.
33. B.C. Kuo, Automatic Control System, Prentice-Hall, New York, 1995.
34. Y.-P. Kuo, M.-T. Liu, and J.-S. Chiou, in: Design of an Evolutionary-Based Fuzzy System, Asian Simulation Conference, 2006.
35. K. Ogata, Modern Control Engineering, second ed., Prentice-Hall, India, 1992.
36. S.-K. Oh, and W. Pedrycz, Genetic optimization-driven multi-layer hybrid fuzzy neural networks, Simulation Modelling Practice and Theory 14 (2006) 597–613.
37. S.-K. Oh, W. Pedrycz, and K.-J. Park, Structural developments of fuzzy systems with the aid of information granulation, Simulation Modelling Practice and Theory 15 (2007) 1292–1309.
38. K.-J. Park, W. Pedrycz, and S.-K. Oh, A genetic approach to modeling fuzzy systems based on information granulation and successive generation-based evolution method, Simulation Modelling Practice and Theory 15 (2007) 1128–1145.
39. K.M. Passino, and S. Yurkovich, Fuzzy Control, Addison-Wesley, Reading, MA, 1998.
40. A. Pollard, Process Control, Heinemann Educational Books, London, 1971.
41. D. Puangdownreong, T. Kulworawanichpong, and S. Sujitjorn, Input weighting optimization for PID controllers based on the adaptive tabu search, 2004 IEEE Region 10 Conference D (2004) 451–454.
42. Al-Aomar Raid, Incorporating robustness into genetic algorithm search of stochastic simulation outputs, Simulation Modelling Practice and Theory 14 (2006) 201–223.
43. X.-J. Wu, X.-J. Zhu, G.-Y. Cao, and H.-Y. Tu, Dynamic modeling of SOFC based on a T–S fuzzy model, Simulation Modelling Practice and Theory 16 (2008) 494–504.
44. N. Saad, and V. Kadirkamanathan, A DES approach for the contextual load modelling of supply chain system for instability analysis, Simulation Modelling Practice and Theory 14 (2006) 541–563.
45. E.N. Sanchez, H.M. Becerra, and C.M. Velez, Combining fuzzy, PID and regulation control for an autonomous mini-helicopter, Information Sciences 177 (2007) 1999–2022.
46. F. van den Bergh, and A.P. Engelbrecht, A study of particle swarm optimization particle trajectories, Information Sciences 176 (2006) 937–971.
47. D.W. Van der Merwe, and A.P. Engelbrecht, Data clustering using particle swarm optimization, Proceedings of the 2003 Congress on Evolutionary Computation (2003) 215–220.
48. Y. Yang, and M. Kamel, Clustering ensemble using swarm intelligence, in: Proceedings of the 2003 IEEE Swarm Intelligence Symposium, SIS’03, 2003, pp.65–71.
49. H. Yoshida, K. Kawata, Y. Fukuyama, and Y. Nakanishi, A particle swarm optimization for reactive power and voltage control considering voltage security assessment, IEEE Transactions on Power Systems 15 (2000) 1232–1239.
50. J.-S. Chiou, and M.-T. Liu, Numerical simulation for Fuzzy-PID controllers and helping EP reproduction with PSO hybrid algorithm, Simulation Modelling Practice and Theory 17 (10) (2009) 1555–1565.
51. R.C. Eberhart, and Y. Shi, Comparing inertia weights and constriction factors in particle swarm optimization, in: Proceedings of the IEEE Congress on Evolutionary Computation, San Diego, USA, 2000, pp. 84–88.
52. L.-L. Li, L. Wang, and L.-L. Liu, An effective hybrid PSOSA strategy for optimization and its application to parameter estimation, Applied Mathematics and Computation 179 (1) (2006) 135–146.
53. M. M. Noel and T. C. Jannett, Simulation of a new hybrid particle swarm 93 optimization algorithm, in Proc 36th Southeastern Symp. Syst. Theory, 2004, pp.150–153.
54. C. F. Juang, A hybrid of genetic algorithm and particle swarm optimization for recurrent network design, IEEE System, Man, Cybern: B 34 (2) (2004) 997–1006.
55. T. Huang, and A. S. Mohan, A Hybrid Boundary Condition for Robust Particle Swarm Optimization, IEEE Antennas and Wireless Propagation Letters 4 (2005)112-117.
56. D. Liu, K. C. Tan, C. K. Goh, and W. K. Ho, A Multiobjective Memetic Algorithm Based on Particle Swarm Optimization, IEEE System, Man, Cybern:B 37 (1) 2007 42-50.
57. A. Beghi, M. Liberati, S. Mezzalira, and S. Peron, Grey-box modeling of a motorcycle shock absorber for virtual prototyping applications, Simulation Modelling Practice and Theory 15 (8) (2007) 894–907.
58. R. Bellman, and R. Kalaba, Dynamic programming and modern control theory (New York: Academic Press Inc., 1965).
59. İkbal Eski, and Şahin Yıldırım, Vibration control of vehicle active suspension system using a new robust neural network control system, Simulation Modelling Practice and Theory 17 (5) (2009) 778–793.
60. K. Gulez, and R. Guclu, CBA-neural network control of a non-linear full vehicle model, Simulation Modelling Practice and Theory 16 (9) (2008) 1163–1176.
61. J. Cao, H. Liu, P. Li, and D. J. Brown, State of the Art in Vehicle Active Suspension Adaptive Control Systems Based on Intelligent Methodologies, IEEE Intelligent Transportation Systems 9 (3) (2008) 392-405.
62. D. Carlos, and Paternina-Arboleda, Tapas K. Das, A multi-agent reinforcement learning approach to obtaining dynamic control policies for stochastic lot scheduling problem, Simulation Modelling Practice and Theory 13 (5) (2005)389–406.
63. J. Z. Feng, J. Li and F. Yu, GA-based PID and fuzzy logic control for active vehicle suspension syetem. International Journal of Automotive Technology 4 (4) (2003) 181-191.64. I. Youn, and A. Hac, Preview control of active suspension with integral action.International Journal of Automotive Technology 7 (5) (2006) 547-554.
65. J.-S. Chiou, and M.-T. Liu, Using Fuzzy Logic Controller and Evolutionary Genetic Algorithm for Automotive Active Suspension System, International Journal of Automotive Technology 10 (6) (2009) 703-710.