Fault Diagnosis for Nonlinear Networked Control Systems Based on T-S Fuzzy Model

Abstract

In this paper, a fault diagnosis method is proposed for nonlinear networked control systems (NCSs) with random delays. First, a two-layer quasi T-S fuzzy model based on probability is presented for the NCSs. Stochastic and nonlinear features of the NCSs are incorporated in the model. Then, based on this model the fuzzy observer and the residual generator are designed to estimate the unmeasurable state and indicate faults. Sufficient Conditions on the stability of the fuzzy observer and the existence of the robust residual generator are presented. Finally, an example is included to show the efficiency of the proposed method. for fault diction were developed. Fang presented a quasi T-S fuzzy model for linear and nonlinear NCSs with random time delays, and analyzed the stability of the closed-loop NCSs (4). Motivated by these works we use a two-layer quasi T-S fuzzy model based on probability to model for nonlinear NCSs with random delays. Based on this model the fuzzy observer and the residual generator are designed to estimate the unmeasurable state and indicate faults. The stability condition for the fuzzy observer and the existing condition of the robust residual generator are presented.