Fault Tolerant Control Design Based on Multiobserver of Sensor Faults for Fuzzy T-S Systems

Abstract

In this paper, the work’s goal is to analyze as well as conceive a sensor fault-tolerant fuzzy T-S controller, based on the feedback of reconstructed states, via the Takagi-Sugeno (T-S) fuzzy model which is influenced by the sensor faults. The control strategy proposed in this work aims at stabilizing the closed loop system, although sensor faults exist, as well as compensating their effects. For this, a multiobserver is designed so as to estimate, at the same time, the system states and the sensor faults. For the design of the sensor fault-tolerant fuzzy T-S controller, we base the analysis and the stability study on the Lyapunov quadratic function and the Linear Matrix Inequalities (LMIs) formulation. Furthermore, the performance optimization technique aims to reduce the derivation impacts of sensor faults on the system states and estimation errors. The multiobserver and the fuzzy T-S controller gains are simultaneously determined by the resolution of the LMI conditions. Finally, the control approach proposed in this work is applied in simulation to a real example which is a manipulator arm operated by a DC motor, so as to show the suggested control technique efficiency.