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：	魏宏宇 Wei, Hong-Yu
論文名稱：	具有干擾項之最佳迭代學習演算控制 Optimal Iterative-Learning Control with Disturbances
指導教授：	蔡聖鴻 Tsai, Sheng-Hong Jason
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2014
畢業學年度：	102
語文別：	英文
論文頁數：	113
中文關鍵詞：	觀測器/卡爾曼濾波器鑑別、迭代學習、誤差補償、數位再設計、錯誤預測
外文關鍵詞：	Observer/Kalman filter identification, iterative learning control, error compensation, digital redesign, error prediction
相關次數：	點閱：133 下載：1
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

本篇論文提出利用最佳誤差迭代學習演算法去處理含有干擾項之已知或未知系統。首先，有時我們很難在含有干擾項的已知系統中得到系統的狀態，因此我們設計了基於預測的數位觀測器去估測系統之狀態並利用最佳誤差迭代學習演算法去處理此項問題。再來，在真實世界裡，有很多未知系統的資訊是非常難去獲得的，因此我們利用觀測器/卡爾曼濾波器鑑別方法去處理此遇到之問題。接著，我們利用數位再設計方法去設計出含有高增益比值的控制器去提升系統的軌跡追蹤能力。其中值得一提的是在利用完觀測器/卡爾曼濾波器鑑別方法鑑別完系統且得到系統模型之後，利用最佳誤差迭代學習演算法我們可以使用較低階的系統模型去控制較高階的真實系統。再者，錯誤預測的觀念也被利用來增強最佳誤差迭代學習演算法，使其追蹤能力更好。最後，我們也將證明最佳誤差迭代學習演算法的容錯能力。

In this thesis, optimal error compensation iterative learning control (OECILC) has been implemented to deal with steady state error from a known or an unknown system with disturbances. First, consider a known system with disturbances. Sometimes it is not easy to obtain the states of system, so we design a prediction-based digital observer to estimate the states from the system and then apply OECILC to deal with this problem. Furthermore, consider there are lots of unknown systems in real world and the information is hard to obtain, then Observer/Kalman filter identification (OKID) is applied to deal with this problem. Next, we use digital redesign to design the controller and use high ratio of Q to R to promote the performance. To be worth mentioning, after OKID, we use lower degree system model from OKID to control higher degree real system with OECILC. In addition, error prediction concept is applied to enhance OECILC and make the performance better. Final, the fault tolerance of OECILC is discussed.

List of Contents
中文摘要	i
Abstract	ii
致謝	iii
Acknowledgement	iv
List of Contents	v
List of Figures	viii

Chapters
1.	Introduction	1
2.	Novel Optimal Error Compensation Iterative Learning Observers and Trackers for the System with Disturbances	5
2.1	A New Linear Quadratic Analog Tracker Design for the System with Disturbances	6
2.2	A New Digital-redesign LQT for the Sampled-data System with Disturbances	10
2.3	A New Iterative Learning LQT for the Repetitive System with Unknown Disturbances	14
2.4	New Prediction-based Digital Observer for the System with Disturbances	18
2.4.1	A new prediction-based digital observer for the system with known disturbances	18
2.4.2	A new optimal error compensation iterative learning observer for the repetitive sampled-data system with unknown disturbances	22
2.5	Extended OECILO and OECILT : From A Given Linear System with Disturbances to a Unknown Nonlinear System with Disturbances and Fault Tolerance	24
2.5.1	Observer/Kalman filter identification	25
2.5.2	Description of the system fault	32
2.5.3	Design procedure	33
3.	Illustrative Examples	35
Example 3.1	An known linear system with known disturbances	35
Case 3.1-1	disturbances-free	37
Case 3.1-2	disturbances d(t) and s(t) are 1	38
Case 3.1-3	disturbances d(t)=cos(t) and s(t)=sin(t)	40
Case 3.1-4	disturbances d(t) and s(t) are saw teeth	42
Case 3.1-5	disturbances d(t) and s(t) are random	44
Example 3.2	An improved observer for the known linear system with known disturbances	47
Case 3.2-1	disturbances-free	47
Case 3.2-2	disturbances d(t) and s(t) are 1	49
Case 3.2-3	disturbances d(t)=cos(t) and s(t)=sin(t)	53
Case 3.2-4	disturbances d(t) and s(t) are saw teeth	57
Case 3.2-5	disturbances d(t) and s(t) are random	61
Example 3.3	OECILC without error prediction for the known linear system with unknown disturbances	66
Case 3.3-1	disturbances d(t) and s(t) are 1 which can be seen as unknown	
			66
Case 3.3-2	disturbances d(t)=cos(t) and s(t)=sin(t) which can be seen as unknown	67
Case 3.3-3	disturbances d(t) and s(t) are saw teeth which can be seen as unknown	68
Case 3.3-4	disturbances d(t) and s(t) are random which can be seen as unknown	69
Example 3.4	OECILC with error prediction for the known linear system with unknown disturbances	70
Case 3.4-1	disturbances d(t) and s(t) are 1 which can be seen as unknown	
			70
Case 3.4-2	disturbances d(t)=cos(t) and s(t)=sin(t) which can be seen as unknown	72
Case 3.4-3	disturbances d(t) and s(t) are saw teeth which can be seen as unknown	74
Case 3.4-4	disturbances d(t) and s(t) are random which can be seen as unknown	75
Example 3.5	OECILE with an improved observer for the known linear system with unknown disturbances	77
Case 3.5-1	disturbances d(t) and s(t) are 1 which can be seen as unknown	77
Case 3.5-2	disturbances d(t)=cos(t) and s(t)=sin(t) which can be seen as unknown	80
Case 3.5-3	disturbances d(t) and s(t) are saw teeth which can be seen as unknown	83
Case 3.5-4	disturbances d(t) and s(t) are random which can be seen as unknown	86
Example 3.6	Fault tolerance of OECILC for the known linear system with unknown disturbances	89
Example 3.7	OKID for the unknown linear system with unknown disturbances and system fault	92
Example 3.8	OKID for the unknown nonlinear system with unknown disturbances and system fault	95
Conclusion	101
References	103
Appendix A	107

                                    

[1] Arimoto, S., S. Kawamura, and F. Miyazaki, “Bettering operation of robots by learning,” Journal of Robotic Systems , vol. 1(2), pp. 123-140, 1984.
[2] Chen, C. H., J. S. H. Tsai, M. J. Lin, S. M. Guo, and L. S. Shieh, “A novel linear quadratic obersver and tracker for the linear sampled-data regular system with a direct feedthrough term: digital redesign approach,” IMA Journal of Mathematical Control and Information 2013.
[3] Chen, F. M., J. S. H. Tsai, Y. T. Liao, S. M. Guo, M. C. Ho, F. Z. Shaw, and L. S. Shieh, “An improvement on the transient response of tracking for the sampled-data system based on an improved PD-type iterative learning control,” Journal of The Franklin Institute, vol. 351, pp. 1130-1150, 2014
[4] Dedieu, H. and M. J. Ogorzalek, “Controlling chaos in Chua’s circuit via sampled inputs,” International Journal of Bifurcation Chaos, vol. 4(2), pp. 447-455, 1994.
[5] Gilbert, E. G., “Controllability and observability in multivariable control systems,” SIAM Journal on Control, vol. 2(1), pp. 128-151, 1963.
[6] Guo, S. M., L. S. Shieh, G Chen, C. F. Lin, “Effective chaotic orbit tracker a prediction based digital redesign approach,” IEEE Transaction on Circuits and System-I, Fundamental Theory and Applications, vol. 47, pp. 1557-1570, 2000.
[7] Guo, S. M. and Z. H. Peng, “An observer-based decentralized tracker for sampled-data systems: an evolutionary programming approach,” International Journal of General Systems, vol. 34, pp.421-449, 2005.
[8] Ho, B. L. and R. E. Kalman, “Effective construction of linear state-variable models from input-output data,” Proceedings of the Third Annual Allerton Conference on Circuit and System Theory, Monticello, IL, pp. 449-459, 1965.
[9] Huang, C. M., J. S. H. Tsai, R. S. Provence, and L. S. Shieh, “The observer-based linear quadratic sub-optimal digital tracker for analog systems with input and state delays,” Optimal Control Applications and Methods, vol. 24(4), pp. 197-236, 2003.
[10] Juang, J. N. and R. S. Pappa, “Effects of noise on modal parameters identified by the eigensystem realization algorithm,” Journal of Guidance, Control, and Dynamics, vol. 3, pp. 294-303, 1986.
[11] Kuo, B.C., Digital Control Systems, Holt, Rinehart and Winston, New York, 1980.
[12] Lewis, F. L. and V. L. Syrmos, Optimal Control. John Wiley & Sons, New Jersey, 1995.
[13] Nasiri, M. R., “A optimal iterative learning control for continuous time system,” IEEE Industrial Electronics, IECON 2006-32nd Annual Conference on,” Paris (2006).
[14] Sheen, I. E., J. S. H. Tsai, and L. S. Shieh, “Optimal digital redesign of continuous-time systems using fractional-order hold,” Optimal Control Applications and Methods, vol. 18, pp. 399-422, 1997.
[15] Sheen, I. E., J. S. H. Tsai, and L. S. Shieh , “Optimal digital redesign of continuous-time system using nonideal sampler and zero-order hold,” Journal of Control Systems and Technology, vol. 5(4), pp. 243-252, 1997.
[16] Sheen, I. E., J. S. H. Tsai, and L. S. Shieh, “Optimal digital redesign of continuous-time systems with input time delay and/or asynchronous sampling,” International Journal of the Franklin Institute, vol. 335B(4), pp. 605-616, 1998.
[17] Shieh, L. S., X. M. Zho, and J. L. Zhang, “Locally optimal-digital redesign of continuous-time systems,” IEEE Transactions Automatic Control, vol. 36, pp. 511-515, 1989.
[18] Shieh, L. S., B. B. Decrocq, and J. L. Zhang, “Optimal digital redesign of cascaded analogue controllers,” Optimal Control Applications and Methods, vol. 12(4), pp. 205-219, 1991.
[19] Tsai, J. S. H. , L. S. Shieh, J. L. Zhang, and N. P. Coleman, “Digital redesign of pseudo-continuous-time suboptimal regulators for large-scale discrete systems,” Control Theory and Advanced Technology, vol. 5(1), pp. 37-65 ,1989.
[20] Tsai, J. S. H., C. T. Wang, and L. S. Shieh, “Model conversion and digital redesign of singular systems,” International Journal of the Franklin Institute, vol. 330(6), pp. 1063-1086, 1993.
[21] Tsai, J. S. H., C. C. Huang, S. M. Guo, and L. S. Shieh, “Continuous to discrete model conversion for the system with a singular system matrix based on matrix sign function,’’ Applied Mathematical Modelling, vol. 35(8), pp. 3893-3904, 2011.
[22] Tsai, J. S. H., Y. Y. Du, W. Z. Zhuang, S. M. Guo, C. W. Chen, and L. S. Shieh, “Optimal anti-windup digital redesign of multi-input multi-output control systems under input constraints,” IET Control Theory Applications, vol. 5(3), pp. 447-464, 2011.
[23] Yen, G. G., and L. W. Ho, “Online multiple-model-based fault diagnosis and accommodation,” IEEE Transactions on Industrial Electronics, vol. 50, pp. 296-312, 2003.

校內：2019-07-04公開
校外：2024-07-04公開

簡易檢索 / 詳目顯示

相關論文