Observer-based Fault-Tolerant Control of an Electro-Hydraulic Actuator with mismatched disturbance

Abstract

This article develops a novel fault-tolerant controller for an electro-hydraulic actuation system in the presence of mismatched disturbance and sensor malfunction. In detail, on the basic of the linear matrix inequality (LMI) approach, a nonlinear unknown input observer (NUIO) is designed to simultaneously estimate the sensor fault and output of position that is not impacted by the external disturbance. Next, a fault detection module is constructed to detect sensor fault and feedback the estimated position to the controller when the fault occurs. For the purpose of approximating and reducing the influence of the mismatched disturbance, a radial basis function neural network (RBFNN) is deployed. The combination of NUIO, RBFNN in a dynamic surface control (DSC) approach is introduced to not only achieve precision tracking control but also guarantee a stable system in the event of sensor fault. Furthermore, the Lyapunov function method is utilized to analyze the stability of the resulting closed-loop system. Finally, simulation results and comparison studies are presented to verify the effectiveness and feasibility of the theoretical claims.