Adaptive fault diagnosis observer design for a class of nonlinear system with model uncertainties

In this paper, we propose an adaptive fault diagnosis observer (AFDO) for a class of nonlinear systems with uncertainties. The AFDO, which estimates the faults in real time, is designed based on Lyapunov analysis. We design the adaptive fault estimation laws for AFDO by using the characteristics of output errors. Moreover, in order to handle the chattering problem, we modify the output error term to saturation term. And we add the e-modification term to handle the parameter drift, which also relaxes the assumption that the information about the upper bound of disturbancees is known prior. By Lyapunov analysis, we prove the uniform ultimately boundedness (UUB) of the system output error and unpredictable faults. Finally, we verify the effectiveness of the proposed method through computer simulations.