Observer-Based Fault Diagnosis and Fault-Tolerant Tracking Control for T-S Fuzzy Uncertain System Affected by Simultaneous Sensor and Actuator Faults

The work presented in this article seeks to address the fault-tolerant tracking control (FTTC) problem for T-S fuzzy uncertain continuous nonlinear systems affected by sensor and actuator faults (SAFs). The T-S fuzzy model is utilized to deal with nonlinearity and uncertainty of the system. Based on a novel fuzzy observer which is constructed to estimate the values of immeasurable states and SAFs at the same time, a novel PI-type fuzzy fault-tolerant control law is mentioned to deal with the effect of SAFs simultaneously. By using Linear Matrix Inequalities (LMIs), the sufficient design conditions are converted to a convex optimization problem. The gains of reference model, observer and fault tolerant control law are obtained easily by solving the LMI conditions represented in the Theorems. The stability of the target system is ensured by applying the quadratic Lyapunov function. Finally, simulations and comparisons are given to show the validity and effectiveness of the proposed approach by two practical examples: Inverted Pendulum on a Cart and Overhead Crane System. For instance, the results presented that the tracking performance was enhanced by 23% when the faults occurred in Case 3 of Example 1, in which the comparisons are carried out with the method in Bouarar et al., (2013). The estimation error of SAFs in all cases were not exceed 2% at the steady stage of simulation.