Faults detection for nonlinear systems with H∞ filter

In this paper, fault detection technique is concerned for nonlinear systems which are described by Takagi-sugeno (T-S) fuzzy parts with measurable premise variables and uncertainty parts. In addition, the real system is likely to contain a fault signal owing to various factors. As a result, it is difficult to stabilize the control system when a fault occurs. With the aid of the Kalman-Yakubovic-Popov (KYP) lemma in finite-frequency domains, a new linear-matrix-inequality-based fault detection method for control systems is obtained, which not only guarantees the stability and reliability for unlinear dynamic systems in fault-free case and faulty cases, but also optimize the fault sensitivity performance index and gain the sensitivity to fault. It should be noted that this method can reduce the complexity that comes from the design of filter and fault detection respectively. Finally, the numeral examples and its simulation results are given to illustrate the effectiveness of the proposed design method.