Estimation/decoupling approach for robust Takagi–Sugeno UIO-based fault reconstruction in nonlinear systems affected by a simultaneous time-varying actuator and sensor faults

Abstract

This paper presents a robust design of Takagi–Sugeno Multiple-Integral Unknown Input Observer (TSMIUIO) for nonlinear systems affected by a simultaneous time-varying actuator and sensor faults, sensor noise and unknown input by using novel estimation/decoupling approach. In comparison to the available literature, the motivation for this approach is to: (1) attain robust fault reconstruction against the effects of bi-directional interaction accompanying the use of extended state observer (ESO) to simultaneously estimate actuator and sensor faults. (2) Guarantee accurate fault reconstruction despite the time behaviour of the fault. (3) Reduce the order of the TSMIUIO by avoiding the use of ESO for estimating all faults. The L2-norm minimisation has been exploited to ensure robust global stability against system disturbances. The design algorithm has been formulated using linear matrix inequality framework. Finally, the effectiveness of the proposal is illustrated by using a single-link flexible joint robot.