臺灣博碩士論文加值系統

摘要


面對可靠的控制系統可靠性的要求愈來愈高，故障診斷系統愈形重要。本論文中採用以模式為基礎設計之未知輸入解耦之觀測器( UIO )設計方法，以克服不可避免的模式不準度而達成穩健之故障偵測及區隔之工作要求。該觀測器係採用特徵結構極點配置之設計方法，其中選定之特徵值對應了指定的故障偵測速度，而特徵向量之選定則影響故障與偵測殘值間之交互作用。
設計上，希望達到故障與殘值間為對角化之關係以達成精確的故障區隔。本論文使用靜態解耦方式讓故障 \ 殘值間之直流增益儘量為”1”，以便估計出故障之大小。文中定義交互作用指標( interaction index )作為成本函數，則數值最佳化方法便可進行以決定適當的特徵向量，以達成故障 \ 殘值間之最低交互作用。
本論文提出之設計步驟透過一垂直升降飛機( VTOL )之動態模擬進行驗證，模擬之結果顯示了本方法之有效性。

Abstract


The increasing demands for reliable control system require improv-
eed design for fault diagnostic system. In this thesis, the model-based unk-
nown input observer ( UIO ) design approach is applied to achieve robust f-
ault detection and isolation ( FDI ) in the face of unavoidable model uncert-
ainties.

The observer is designed with eigenstructrue assignment method s-
uch that eigenvalues of the design specify the fault detecting speed, while t-
he selecting eigenvectors affect interactions between fault and residuals. It is desired that the fault \ residual map is diagonalized to achieve precise fa-
ult isolation. We adopt the static decoupling method, in this research, to ac-
hieve the D.C. gain of the fault \ residual map approach to identity so as to estimate the fault size. An interaction index is defined and is applied as the cost function so that numerical optimization procedure can be proceed to d-
etermine the proper eigenvectors in which the least interaction of fault \ re-
sidual map can be achieved.

The proposed design procedure is verified through simulation studi-
es performed on a vertical takeoff and landing ( VTOL ) aircraft in the vert-
ical plane. The results of the simulation demonstrate the effectiveness of th-
e proposed design.

穩健式UIO之低交互作用的設計
Low Interaction of Robust UIO Design
目錄


abstract ……………………………………………..………………………I
中文摘要……………………..…………………….……….……..……... II
目錄……………………………..…………………………..…..……..….III
圖目錄…………………………...………………………….………..…....V
表目錄………………….…………………..………..…………………..VII
符號說明……………………………………………………..…….….. VIII

第一章　緒論
1.1 研究背景與發展概況…………………………………………….1
1.2 研究動機與目的………………………………………………….2
1.3 問題描述及研究方法…………………………………………….3
1.4 論文架構………………………………………………………….4

第二章　UIO基礎下的殘值產生器
2.1 故障診斷系統( Fault Diagnosis Supervision )簡介……………..5
2.2 穩健式殘值的產生( Residual Generation )……………….……..7
2.3 UIO的設計…………………………………………..………….10

第三章　穩健式UIO之低交互作用的設計
3.1　交互作用( Interaction )的分析……………………………………14
3.2　 不變子空間( invariant subspaces )之特徵結構極點配置………17
3.3 設計UIO最佳化過程………………………………………….18

第四章　模擬結果與分析
4.1 數值模擬與驗證…………………………………………………23
4.2 致動器FDI系統……………………………….………………...25
4.3 感測器FDI系統…………………………………………………29

第五章　結論與建議
5.1 結論……………………………………………………………….35
5.2 建議……………………………………………………….………36

參考文獻………..……………………………………………….………37

參考文獻

[1] Watanabe, K., and Himmelblau, D. M.(1987) Instrument fault detec-
tion in systems with uncertainties. Internalional Journal of Systems Science, 13, 137-158.

[2] Viswannadham, N., and Srichander, R.( 1987 ) Fault detection using
undnown input observers. Control-theory and Advanced Technology, 3, 91-101.

[3] Frank, P. M., and W nnenberg, J.(1989) Robust fault diagnosis usi-
ng unknown input observer scheme. Fault Diagnosis in Dynamic S-
ystems: Theory and Application , edited by R. J. Patten, P. M. Frank and R. N. Clark ( Englewood Cliffs, New Jersey, U.S.A.: Prentice
Hall ), Chapter 3, pp.47-98.

[4] Patten, R. J., Frank, P. M., and Clark, R. N.,(editors)(1989) Fault Di-
agnosis in Dynamic Systems: Theory and Application ( Englewood
Cliffs, New Jersey, U.S.A.: Prentice Hall ).

[5] Saif, M., and Guan, Y.(1993) A new approach to robust fault detect-
ion and identification. IEEE Transactions on Aerospace and Electr-
onic Systems, 29, 685-695.

[6] Frank, P. M.(1987) Fault diagnosis in dynamic systems via state esti-
mation－a survey. System Fault Diagnostics, Reliability and Relat-
ed Knowledge-Based Approach, Vol. 1, edited by S. G. Tzafestas, M. G. Singh and G. Schmidt ( Dordrecht, The Netherlands: Reidel ), pp. 35-98; (1994)Enhancement of robustness in observer-based fault de
tection. International Journal of Control, 59, 995-981.

[7] Chang, S. K., Hsu, P. L., and Lin, K. L.(1995) A parametric transfer matrix approach to fault identification filter design and threshold se-
lection. International Journal of Systems Science, 26, 741-754.

[8] Patten, R. J., and Kangethe, S. M.(1989) Robust fault diagnosis us-
ing eigenstructrue assignment of observers. Fault Diagnosis in D-
ynamic Systems: Theory and Application, edited by R. J. Patten, P. M. Frank and R. N. Clark ( Englewood Cliffs, New Jersey, U.S.A.: Prentice Hall ), Chapter 4, pp. 99-154.

[9] Yang, F., and Wilde, R. W.(1988) Observers for linear systems with unknown inputs. IEEE Transaction on Automatic Control, 33, 677
-681.

[10] Campbell, S. L., and Meyer Jr., C. C.(1979). Generalized Inverses of Linear Transformation ( London, U.K.: Pitman ).

[11] Chen Jie and Patton R. J.(1999). Robust Model-Based Fault Diagno-
sis for Dynamic Systems, Kluwer academic publishers.(p.77)

[12] Y. Tai, K.Yoshizawa, and H. Okubo .(2002) . Robust fault detection of smart structure control systems with model uncertainty, SICE. 02-0365

[13] Davison, E. J.(1970). Interaction index for multivariable control sys-
tems, Proc. IEE, 177: 459-462

[14] Sinha.P. K. (1985).Multivariable Control An Introduction, Marcel Dekker, Inc. New York and Basel.

[15] Jeffery B. Burl (1999). Linear Optimal Control and Meth-
od, Addison-Wesley.

[16] Liu, G. P. and Patton, R. J. (1998). Eigenstructure Assignment for C-
ontrol System Design, John Wiley & Sons Ltd, Chichester.

[17] Narendra, K. and Tripathi, S. S. Identification and optimization of ai-
rcraft dynamics. AIAA Journal of Aircraft, 1973, 10, 193-199

[18] Hwang, D. S., Chang, S. K. and Hsu, P. L. (1997). A practical design for a robust fault detection and isolation system, Int. J. Sys. Sci. 28(
3):265-275

[19] L tkepohl H. (1996), Handbook of Matrices, John Wiley & Sons Ltd, Chichester.

[20] Roger A. Horn and Charles R. Johnson (1990). Matrix Analysis, Ca-
mbridge University Press.

[21] Robert E. Skelton, Tetsuya Iwasaki and Karolos M. Grigoriadis (199
8). A Unified Algebraic Approach to Linear Control Design, Taylor & Francis.

[22] Steffen Leonhardt, Norbert M ller, and Rolf Isermann (1999). Met-
hods for engine supervision and control based on cylinder pressure information, IEEE Trans. on Mechatronics, 4 (3): 235-246.

[23] Shen. Li-Cheng and Hsu, Pau-Lo (1998). Robust design of fault isol-
ation observers, Automatic 34(11): 1421-1429.

[24] Silvio Simani, Cesare Fantuzzi and Ron J. Patton (2002). Model-ba-
sed Fault Diagnosis in Dynamic Systems Using Identification Tech-
niques , Springer.

[25] Chiang L. H., E. L. Russell and R. D. Braatz (2000). Fault Detection and Diagnosis in Industrial Systems, Springer.

[26] Kautsky, J. and N.K. Nichols (1985). Robust pole assignment in lin-
ear state feedback, Int. J. Control, 41(5): 1129-1155.

[27] Tsui Chia-Chi (1996). A new approach to unknown input observers, IEEE Trans. Automat. Contr. 41(3): 464-468.

[28] Dale E. Seborg, Thomas F. Edgar and Duncan A. Mellichamp(1989). Process Dynamics and Control, John Wiley & Sons: 465-467.