In this thesis, the main contributions can be summarized as follows: Robust Fault Diagnosis for a Class of Nonlinear Time-delayed Systems: A robust actuator fault diagnosis approach is developed for a class of nonlinear systems with time-varying delays. Both detection and diagnostic observers are established, where the Lyapunov stability theory is used to obtain the required adaptive tuning rules for the estimation of the nonlinear actuator faults. This has led to stable observers for both fault detection and diagnosis. Integrated Fault Tolerant Control of a Class of Nonlinear Time-delayed Systems: An integrated fault tolerant control approach is proposed for a class of nonlinear systems with time-varying delays. An extra control law is added to the system after a fault has happened. A nonlinear adaptive iterative learning fault diagnosis observer is designed. The fault tolerant control is also obtained via the adaptive tuning of the extra control law. The stability of the fault diagnosis and tolerant control can be guaranteed. Fault Tolerant Control for Collaborative Systems and Sequential Continuous Systems: A new fault tolerant control method is proposed for a class of dynamic systems including two subsystems. When one subsystem, incapable of self-repairing, is subjected to faults, the other subsystem is used to accommodate faults and compensate the influences onto the total system, leading to a fault tolerant control of the whole system. This fault tolerant control approach has been generalized to sequential continuous Systems. Fault Diagnosis and Tolerant Control for Non-Gaussian Stochastic Systems: A nonlinear adaptive observer-based fault diagnosis algorithm is proposed for non-Gaussian stochastic distribution systems based on the rational square-root B-spline model approximation. Through the controller reconfiguration, a good output PDF tracking can still be realized when fault occurs, leading to the fault tolerant control of the non-Gaussian stochastic distribution systems. Fault Diagnosis for a Class of Singular Lipschitz Nonlinear Systems: A new fault diagnosis algorithm is developed for a class of singular Lipschitz nonlinear systems. A regular nonlinear adaptive diagnostic observer is established, where the adaptive tuning rule for estimation of fault is obtained under certain conditions. The stability of the diagnostic observation error system is also proven. Fault Diagnosis and Tolerant Control for Singular Linear Stochastic Distribution Systems: Suitable coordinate conversion has been introduced to transform singular linear stochastic distribution systems into differential-algebraic stochastic distribution systems. A fault diagnosis approach is presented based on an adaptive observer. Through the controller reconfiguration using the fault estimation information, a good output PDF tracking for the pre-specified PDF can still be realized when fault occurs.
修改评论