An adaptive observer-based fault detection and diagnosis for nonlinear systems with sensor and actuator faults

A novel fault diagnose method is presented for the nonlinear systems with coupled fault and disturbance via adaptive observer technique. Residual matrix is used in this paper to describe the values of unknown inputs. Instead of adding robustness on the estimate states, this novel observer structure focuses on eliminating the disturbance signal on the residual matrix by using vector unitization so that the residual matrix can be approached precisely. The adaptive law is obtained by using Lyapunov stability theory. Finally, an application on second order system is included in the end of this paper to show the effectiveness of the proposed strategy.