A novel double-level observer-based fault estimation for Takagi–Sugeno fuzzy systems with unknown nonlinear dynamics

Abstract

In this paper we investigate the fault estimation problem against local unknown nonlinear dynamics, sensor and actuator faults for a class of Takagi–Sugeno (T-S) fuzzy systems. In addition, the exogenous disturbances and measurement noise are considered, which are presented in the operation of the systems and are various and independent of the systems. A novel double-level observer is designed to estimate the system states and faults. Compared with the current research results, the proposed observer has a wider range of application. By designing a fuzzy augmented system and a Kalman filter as the first-level observer, the estimations of system states, sensor faults and actuator faults can be obtained simultaneously. The second-level observer can estimate the unknown nonlinear dynamic function by establishing generalized fuzzy hyperbolic model. The robust stability of the estimation error systems is considered by H∞ performance. Finally, three simulation examples are provided to demonstrate the effectiveness of the proposed fault estimation method.