迴路整型(loop shaping)技術為強健控制領域中的熱門研究主題， 而其中之一即為指定頻段H∞控制技術，其可應用廣義KYP (generalized Kalman-Yakubovich -Popov，GKYP)引理逕行設計，而毋需引用權重函數。相對地，指定頻段H2控制方法因受限於缺乏相關數學工具，並無這樣的直接設計方法。因此，引入權重函數的輔助設計仍為目前主要的近似解法。然而，權重函數的引入會造成兩個問題，其一為控制器階數遽增，增添了硬體實現之複雜度；其二會在擴增系統內增加無法移動的極點，進而可能會影響閉回路系統(不含權重函數)極點之配置。基於上述分析，本論文聚焦於傳統含權重函數之H2迴路整型技術的補強之道，特別是控制器階數以及極點是否可任意配置之問題。
本論文針對前述問題，提出三種設計方法。在設計條件有解的情況下，本論文的設計方法可使控制器的階數介於受控體階數與傳統方法所得階數(亦即受控體與權重函數階數之總和)之間。而在閉迴路極點的問題上，由於權重函數亦為設計的一部份，在可穩定與可偵測的前提之下，本論文所提方法可使閉迴路極點配置於指定區域(方法一)或任意指定位置(方法二、三)。本論文所提出的設計條件均為線性矩陣不等式，可運用現有數值軟體模擬迅速求解。最後，本論文針對一低頻雜訊問題，應用本文方法進行設計與模擬，結果證實我們所提出的方法的確有效。

Loop shaping technique is a popular research topic in robust control, and one of them is finite frequency H∞ control technique, in which the generalized Kalman-Yakubovich-Popov (GKYP) lemma can be applied to synthesize controllers without introducing weighting functions. On the contrary there does not exist such a direct design method for finite frequency H2 control problems because there lacks mathematical tools like the GKYP lemma. Therefore, introducing the weighting functions into the design remains to be one of the major approaches that provide approximate solutions to the problems. However, introducing the weighting functions causes two problems. First, it increases the order of controllers, which in turn increases the implementation complexity in hardware. Second, the augmented system contains the poles of the weighting functions that can't be moved via any controller design. This could possibly affect the closed-loop poles design. Based on the above analysis, this thesis focuses on the reinforcement of the traditional weighted H2 loop shaping technique, in particular on the problems of controller order and closed-loop pole placement.
In this thesis three kinds of design methods are presented. When the solvability conditions are feasible, the order of controllers can be within two numbers, the order of the generalized plant without containing the weights and that of the generalized plant containing the weights. As for the pole placement problem, as the choice of weighting functions is a part of design, it will be shown that, under the stabilizability and detectability assumptions on the plant (without containing the weights), regional pole placement is possible by Method 1 and arbitrary pole placement is possible via Methods 2,3. The solvability conditions for computing the controllers are in terms of linear matrix inequalities (LMIs), which can be efficiently solved by computer software. Finally, the proposed methods are applied to a low-frequency noise rejection problem. The simulation results demonstrate the effectiveness of the proposed methods.

目 錄

中文摘要.....................................I
英文摘要.....................................II
目錄........................................IV 
圖目錄.......................................VI
表目錄......................................VII
第一章 緒論..................................1
	1.1文獻回顧與研究動機..................1
	1.2論文架構..........................3
第二章 背景知識...............................4
第三章 基於多目標控制方法之降階控制器設計..........9
	3.1前言.............................9
	3.2 降階控制器設計:方法一..............10
              3.2.1  問題敘述................10
              3.2.2  控制器降階設計...........11
第四章 基於尤拉參數式之降階控制器設計.............16
	4.1 前言............................16
	4.2 降階控制器設計:方法二..............17
              4.2.1  問題敘述................17
              4.2.2  控制器降階設計...........18
              4.2.3  閉迴路極點配置條件........20
              4.2.4  H2性能合成條件...........22
	      4.3 降階控制器設計:方法三.........27
第五章 數值模擬................................33
第六章 結論與未來研究方向........................47
參考文獻......................................49
附錄..........................................51 

圖目錄

圖2.1	P-K架構.................................7
圖3.1	系統架構I..............................10
圖3.2	系統架構II.............................12
圖4.1	系統架構I..............................17
圖4.2	系統架構II.............................19
圖4.3	系統架構III............................22
圖5.1	控制系統方塊圖..........................33
圖5.2	低頻高斯干擾之產生方式....................35
圖5.3	低頻有界能量之產生方式....................36
圖5.4	Tzw與W之振幅響應圖.......................37
圖5.5	低頻高斯干擾抑制效果......................38
圖5.6	雜訊抑制效果比較(noise1)..................39
圖5.7	低頻有界能量干擾抑制效果...................40
圖5.8	雜訊抑制效果比較(noise2)..................41
圖5.9	Tzw1與Tzw2之振幅響應圖....................45
圖5.10	阻振控制的輸出特性.........................46

表目錄

表5.1	閉迴路極點配置.............................35
表5.2	閉迴路極點位置.............................42
表5.3	四種方法的相關數據..........................43
表5.4	權重閉迴路v2範數之近似值 .....................43

[1]K. Zhou and J. C. Doyle, Essentials of Robust Control, Prentice Hall, 1998.
[2]C.W. Scherer, P. Gahinet and M. Chilali, “Multiobjective output-feedback control via LMI optimization,” IEEE Trans. Aut. Control, vol. 42, 1997, pp. 896-911.
[3]C.W. Scherer, “An efficient solution to multi-objective control problems with LMI objectives,” Systems & Control Letters, vol. 40, 2000, pp. 43-57.
[4]M.C. de Oliveira, J.C. Geromel and J. Bernussou, “Extend   and  norm characterizations and controller parameterizations for discrete-time systems,” Int. J.Control, vol. 75, no. 9, 2002, pp. 666-679.
[5]T.S. Lee, S.J. Chiang and J.M. Chang, “  loop-shaping controller design for the single-phase ups inverters,” IEEE Trans. on Power Electronics, vol. 16, no. 4, July 2001, pp. 473-481.
[6]K. Nam, S. Oh and Y. Hori, “Robust yaw stability control for electric vehicles based on steering angle-disturbance observer and tracking control design,” IECON-36th Annual Conf. on IEEE Industrial Electronics Society, Nov. 2010, pp. 1943-1948.
[7]W. Sun, H. Gao, and O. Kaynak, “Finite frequency   control for vehicle active suspension systems,” IEEE Trans. on Control Systems Technology, vol.19, no. 2, Mar. 2011, pp. 416-422.
[8]R. de Castro, R.E. Araujo, J.P. F.Trovao, P.G. Pereirinha, P. Melo and D. Freitas, “Robust dc-link control in EVs with multiple energy storage systems,” IEEE Trans on Vehicular Technology, vol. 61, no. 8, Oct. 2012, pp. 3553-3565.
[9]T. Iwasaki and S. Hara, “Generalized KYP lemma: unified frequency domain inequalities with design applications,” IEEE Trans. Autom. Control, vol. 50, no. 1, 2005, pp. 41-59.
[10]T. Iwasaki and S. Hara, “Feedback control synthesis of multiple frequency domain specifications via generalized KYP lemma,” International Journal of Robust and Nonlinear Control, vol. 17, no. 5, 2007, pp. 415-434.
[11]P. Gahinet, A. Nemirovski, A. J. Laub and M. Chilali, “LMI control toolbox user's guide,” The Mathworks Inc., USA, 1995.
[12]Chi-Tsong Chen, Linear System Theory and Design third edition, Oxford University Press, 1999.