Direct-Drive Wind Turbine Fault Diagnosis Based on Logistic Regression

Abstract

In this paper, we proposed a fault diagnosis method based on logistic regression model through the analysis of direct-drive wind power generation set under five conditions, normal condition, wind wheel mass imbalance, wind wheel aerodynamic imbalance, yaw and blade break. In the training process of logistic regression, we created a characteristic parameter matrix and set the probability value, then we calculated the logic regression parameter matrix using the method of Newton-Raphson to diagnose the state of wind turbine.