Adaptive Actuator Fault Estimation of Stochastic Systems with Output Measurement Error

This study investigates adaptive fault estimation for non-Gaussian stochastic systems with actuator faults and output measurement error. After fuzzy linearization for the singular systems, the output probability density function is generated by rational square root B-splines. To deal with the intermittent multiple actuator faults, an improved adaptive fault estimation is proposed: A compensation function can directly reconstruct the algorithm to mask the output measurement error; then, actuator fault estimation accurately tracks the multiple actuator faults. A novel variable parameter algorithm that mimics animal predation behavior is designed and incorporated into learning rates, making the controller more sensitive to the incipient deviations in actuator faults. Finally Lyapunov function proves the stability, simulation verifies the effectiveness of the method.