Fault Detection of WECSs With a Delayed Input and an Unknown Part Based on SVO

This paper studies an improved set-valued observer (SVO) for a complex model of a wind energy conversion system (WECS). To fit reality, an unknown function is used to describe the aerodynamic model, and a delay of the input signal is added to simulate the actual situation. The conservative SVO estimates the state error convergence within an adjustable interval. The extended Lipschitz condition, set-induced theory, and uniformly ultimately boundedness guarantees the convergence of errors. Based on the SVO, a fault detection system and corresponding fault detection strategies for common faults of WECSs are proposed. By adjusting the gain matrix L, the SVO shows excellent state tracking performance. The simulation results show that the proposed fault detection system can provide rapid and accurate fault detection.