Fault Detection For Delta Operator Formulated Systems With Time Delay

As delay exists in many industrial processes, fault detection research on time-delay systems has significant meaning. At present, fault detection design either concentrates on continuous-time systems and discrete-time systems respectively or discretizes the system with traditional shift operator. By using traditional shift operator, most of the poles are located in the stable boundary when the sampling rate is high. However, the problem can be well solved with Delta operator, and fault detection for continuous time-delay systems and discrete time-delay systems can be considered in the unified Delta operator framework. Therefore, the analysis of Delta operator based time-delay systems is very important.In the thesis, the research background and technologies of fault detection for time-delay systems are reviewed at the beginning. Then, the Delta operator based fault detection filter design for time-delay systems and uncertain systems is discussed. The major contributions of the thesis are as follows:First, Delta operator formulated fault detection robust filter for time-delay systems is given based on the traditional shift operator design. Besides, the sufficient conditions for the existence of the robust fault detection filter and its explicit expressions are developed. The simulation results show the effectiveness of the proposed approach.Second, fault detection method of uncertain Delta operator systems is established. By using H∞control theory, the problem of fault detection robust filter design can be reduced to H∞filtering. Besides, the method proposed guarantees the high sensitivity of the residual signal to the fault signal and the robustness of the uncertain systems. With Matlab LMI Control Toolbox and linear matrix inequalities (LMIs), the optimal robust filter for Delta operator based systems is obtained.