The Diagnosis Method for Converter Fault of the Variable Speed Wind Turbine Based on the Neural Networks

Abstract

Because the converter circuit model of the doubly-fed wind turbine has strong nonlinear, it is difficult to diagnose fault online usually. In this paper, using the nonlinear mapping characteristic of the neural networks, a BP neural networks fault diagnosis method which bases on wave direct analysis is proposed, which achieves the wind turbine converter circuit fault diagnosis online. The actual run results show the method is effective.