Bayesian regularized artificial neural network for fault detection and isolation in wind turbine

This paper proposes an intelligent scheme based on Bayesian artificial neural network (BNN) for fault detection and isolation (FDI) in variable speed wind turbine. The proposed scheme is based on data-driven fault detection method. The main idea is the use of a certain number of BNN classifiers to deals with different types of faults affecting the wind turbine. Different parts of the process were investigated including actuators and sensors faults. The simulation results show the best performances of the proposed approach.