Wind turbines protection improvement using artificial neural networks

Wind turbines influence the short circuit current in a way, which changes the conventional protection methods of the distribution networks. One of the effects wind turbines is the sudden drop of short circuit current, in a small time duration after the fault occurrences, as the value of fault current reduces to a value below the pickup current of overcurrent protection. In this study, the mentioned problem is discussed and moreover, a novel fault protection technique, based upon shifting the function curve of the relay is introduced. The main idea behind this novel scheme is reducing the amount of pickup current precisely, using an artificial neural network. In order to investigate the feasibility of the proposed scheme a wind power generation system, connected to a distribution network is modeled and investigated in MATLAB/Simulink environment, and the results indicate that the novel technique possesses good fault detection abilities.