Fault Tolerant Control for Non Linear Uncertain Takagi-Sugeno Fuzzy Systems Based on Proportional Integral Sliding Mode Observer

Author(s):  
Imen Youssef ◽  
Atef Khedher ◽  
Kamel Ben Othman
Author(s):  
Riadh Hmidi ◽  
Ali Ben Brahim ◽  
Slim Dhahri ◽  
Fayçal Ben Hmida ◽  
Anis Sellami

This paper proposes fault-tolerant control design for uncertain nonlinear systems described under Takagi-Sugeno fuzzy systems with local nonlinear models that satisfy the Lipschitz condition. First, by transforming sensor faults as ‘pseudo-actuator’ faults, an adaptive sliding mode observer is designed in order to simultaneously estimate system states, actuator and sensor faults despite the presence of norm-bounded uncertainties. Second, an adaptive sliding mode controller is suggested to provide a solution to stabilize the closed-loop system, even in the event of simultaneous occurrence of faults in actuators and sensors. Next, the main objective of the fault-tolerant control strategy is to compensate for the effects of fault based on the feedback information. Therefore, using the LMI optimization method, sufficient conditions are developed with [Formula: see text] to calculate the gains of the observer and the controller. Then, particular attention is paid to the simultaneous maximization, by convex multi-objective optimization, of the Lipschitz nonlinear constant in Takagi-Sugeno fuzzy modelling and uncertainties attenuation level. The results of the simulation illustrate the effectiveness of our fault-tolerant control approach using a nonlinear inverted pendulum with a cart system.


In this paper, the problems of fault estimation and fault-tolerant control for Takagi-Sugeno fuzzy system affected by simultaneous actuator faults, sensor faults and external disturbances are investigated. Firstly, an adaptive fuzzy sliding-mode observer is designed to simultaneously estimate system states and both actuator and sensor faults. Then, based on the online estimation information, a static output feedback fault-tolerant controller is designed to compensate for the effect of faults and to stabilize the closed-loop system. Moreover, sufficient conditions for the existence of the proposed observer and controller with an H∞ performance are derived based on Lyapunov stability theory and expressed in terms of linear matrix inequalities. Finally, a nonlinear inverted pendulum with cart system application is given illustrate the validity of the proposed method.


Sign in / Sign up

Export Citation Format

Share Document