Wind turbine simulator fault diagnosis via fuzzy modelling and identification techniques

For improving the safety and the reliability of wind turbine installations, the earliest and fastest fault detection and isolation are highly required, since it could be used also for accommodation purpose. Modern wind turbines consist of several important subsystems, which can be affected by malfunctions regarding actuators, sensors, and components. From the turbine control point-of-view they are extremely important since provide the actuation signals, the main functions, as well as the measurements. In this paper, a fault diagnosis scheme based on the identification of fuzzy models is described, in order to detect and isolate these faults in the most efficient way, in order also to improve the energy cost, the production rate, and reduce the operation and maintenance operations. Fuzzy systems are proposed here since the model under investigation is nonlinear, whilst the wind speed measurement is uncertain since it depends on the rotor plane wind turbulence effects. These fuzzy models are described as Takagi–Sugeno prototypes, whose parameters are estimated from the wind turbine measurements. The fault diagnosis methodology is thus developed using these fuzzy models, which are exploited as residual generators. The wind turbine simulator is finally employed for the validation of the obtained performances.