Based On Ts Fuzzy Model Fault Diagnosis Method

With the developments of the science and technology, there is an increasing demand for the equipments and controlled systems to be more safe, reliable and effective. Study on fault diagnosis (including fault detection and isolation, FDI) for the complex systems has been receiving more and more attention and has become one of the active topics. Especially; the fault diagnosis for linear systems has already been studied intensively in recent years, and a large amount of achievements on model-based fault diagnosis has been obtained. However, most of the actual systems in the real world are nonlinear systems in some sense. In recent years, much attention has being paid on the FDI for nonlinear systems. To author's best knowledge, however, only a few researches have been carried on this. The research on the problem of FDI for nonlinear systems is not only theoretically interesting and challenging, but also very important in practical applications.The main purpose of this thesis is to investigate the FDI problem for a class of nonlinear systems described by Takagi-Sugeno (T-S) fuzzy models. The main context consists of:First is a brief introduction about the development and situation of the considered problem. Then, two of the main methods of FDI, namely observer based fault detection and parity space based FDI, will be discussed. Simulation research on an inverted pendulum system is presented also.The problem of fault estimation is dealt with for a class of T-S fuzzy systems.The parity space based FDI approach for general linear systems is modified to the design of residual generator for T-S fuzzy systems. A new performance index is introduced, which could increase the robustness of the residual to disturbance and meanwhile minimize the error between the residual and fault. A sufficient condition is derived and further the solution of parity matrix can be obtained by solving linear matrix inequalities (LMIs).The problem of fault detection for general nonlinear systems described by T-S fuzzy model is investigated. A parity space based fault detection residual...