G.G. Yen and L.-W. Ho
Fault tolerant control, discrete Lyapunon stability, fault diagnosis, catastrophic failure
Prompted by the increasing demands in system reliability and availability, fault diagnosis and accommodation has quickly become one of the most active research areas in the intelligent control community. Yet, the online fault accommodation controls for the system under catastrophic failures that result in drastic changes and/or changing of dynamic structure are still unsolved. In this work, a suffcient condition for system online stability with changes of dynamic structure under catastrophic failures has been derived based upon discrete-time Lyapunov stability theory. The theoretical analysis indicates that the online control problem can be solved without complete realization of the system dynamics given satisfaction of specific assumptions. An online fault accommodation control framework is proposed to deal with the desired trajectory-tracking problem for systems suffering from various abrupt and unanticipated catastrophic failures. Once a fault is detected, an artificial neural network is suggested as the online estimator to approximate the most recent behaviour of the unknown system failure dynamics. Effective control signals to accommodate the dynamic failures are then computed through the realization of the estimator for the unknown failures based only upon the partially available information of the faults. Extensive simulation studies have been completed to validate the proposed online control architecture under various multiple failures, noisy environments, and false alarm scenarios. The simulations show encouraging results and demonstrate the effectiveness of the proposed control methodology for unanticipated and unknown time-varying failures in online situations based solely upon insufficient information about the system dynamics.
Important Links:
Go Back
