Study On Fault-Tolerant Control Of Nonlinear System With Actuator Fault

As the technological system of modern engineering gets more complicated, security and reliability of the system has drawn more and more attention, and fault-tolerant control (FTC) arises at the right moment. The basic function of an FTC system is that it can guarantee satisfactory performance not only when all the components in the system are healthy but also faulty. UP to now, many results have been achieved on FTC for linear systems. However, for nonlinear systems the results are limited. This paper will concentrate on the nonlinear systematic fault tolerant control.First of all, the thesis has explained the designing strategy of FTC, which is categorized into two types:redundancy design and fault compensation. In the first type, a certain degree of redundancy is introduced to controller. The resulting closed-loop system is stable not only when all actuators are operational but also only when one actuator experiences fault. In the second type, the on-line FTC scheme is developed based on the fuzzy logic system. The closed-loop stability is established based on Lyapunov’s sense. The resulting closed-loop system is stable by using the corrective control law. Compared with the existing results, the fault is accommodated by the fault compensator. The updating control law is so derived that all the parameters of the closed-loop system are bounded. In nonlinear system fault detection, a new method of minimum residual threshold for fault detection and diagnosis was proposed. Through diffeomorphisms, then using sliding mode parameter identification method get robust control algorithm of actuators to adjust the uncertain nonlinear systems fault. This algorithm does not need to the implementation of the fault parameter identification and need not to derivate output of nonlinear system, or requirements of full rank conditions. The controller is a simplified model of the controller, which is a first order stable system instead of a higher order tracking problem, by introducing sliding mode control to offset or slow interference between the actual dynamic and malfunction caused by the dynamic.The advantage of redundancy design is that it also can be done without knowing the position and magnitude of the fault and without fault detection and isolation (FDI), which reduces the computing time and avoids the false alarms, non-detection and delay due to FDI. The advantage of fault compensation is the dynamics caused by faults that can be accommodated online.Second, with regard to a nonlinear system with states unknown, making use of fuzzy theory and designing corresponding controller can help to ensure the whole system stability and local convergence function. The way to estimate the nonlinear system fault is to design an adaptive fuzzy sliding-mode controller by a fast adaptive fault estimate (FAFE) arithmetic, which will provide good regulation when the system faults. The fuzzy logical system can be directly used as a controller according to corresponding fuzzy control rules.Finally, the thesis has made a conclusion about the study achievement and deficiency and then put forward some ideas and expectations for more nonlinear systematic fault-tolerant control research.