Fuzzy model-based faults diagnosis of the wind turbine benchmark

Abstract Modern Wind turbines are complex systems which can be affected by malfunctions, regarding actuators, sensors, and components. An early faults detection and isolation are then highly required. For this aim, in the present paper, a robust fault detection and isolation (FDI) scheme is developed for a 4.8 MW wind turbine described via Takagi–Sugeno (T–S) multiple models. The FDI method is developed by using Fuzzy sliding mode observer as residual generators. Regarding the evaluation task, a set of pre-defined thresholds designed to indicate the occurrence of faults. Following that, a bank of residual generators is employed appropriately to determine fault type and location. Finally, the wind turbine benchmark is used to evaluate the performances of the proposed method against a set of realistic fault scenarios.