Title :
Design of proportional-integral reconfigurable control systems via eigenstructure assignment
Author :
Zhang, Youmin ; Jiang, Jin
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Western Ontario, London, Ont., Canada
Abstract :
In this paper, an integrated design of fault detection, diagnosis and reconfigurable control for multi-input multi-output (MIMO) systems is proposed. The scheme uses proportional-integral (PI) control structure in the reconfigurable controller so as to recover both the dynamic and steady-state performance of the pre-fault system and to reject unknown constant disturbances. A singular value decomposition (SVD) based eigenstructure assignment (EA) technique is developed to achieve online automatic redesign of the controller. Fault diagnosis and controller reconfiguration mechanisms are carried out using statistical hypothesis tests based on the information from a two-stage adaptive Kalman filter. To achieve improved reconfiguration performance, a multiple reconfiguration controller design scheme is exploited. The proposed approach can deal with: abrupt and incipient faults; total and partial faults; single, multiple and consecutive faults with different type of reference inputs, and with the ability to detect and compensate for unanticipated actuator faults which are described by loss of control effectiveness. The effectiveness of the approach has been demonstrated via two simulation examples
Keywords :
MIMO systems; adaptive Kalman filters; compensation; control system synthesis; eigenstructure assignment; fault diagnosis; filtering theory; multivariable control systems; statistical analysis; two-term control; EA; MIMO systems; PI control structure; SVD-based eigenstructure assignment; abrupt faults; consecutive faults; dynamic performance; fault diagnosis; incipient faults; multi-input multi-output systems; multiple faults; online automatic controller redesign; partial faults; proportional-integral reconfigurable control system design; singular value decomposition; statistical hypothesis tests; steady-state performance; two-stage adaptive Kalman filter; unanticipated actuator fault compensation; unanticipated actuator fault detection; unknown constant disturbance rejection; Automatic control; Control systems; Fault detection; Fault diagnosis; MIMO; Pi control; Proportional control; Singular value decomposition; Steady-state; Testing;
Conference_Titel :
American Control Conference, 2000. Proceedings of the 2000
Conference_Location :
Chicago, IL
Print_ISBN :
0-7803-5519-9
DOI :
10.1109/ACC.2000.876918