In order to reduce the major losses and potential safety hazards caused by the failure of the wind turbine drive chain and ensure the healthy and stable operation of the unit, a fault diagnosis method based on SVM improved by difference evolution(DE) algorithm was proposed for the fault diagnosis of the wind turbine. The original data were processed by using EEMD method to extract effective fault features and realize signal-to-noise separation; the key parameters of SVM were optimized by using difference evolution algorithm to improve the generalization ability and prediction accuracy of the model; the trained model was used for fault diagnosis. The results show that the proposed method is accurate and effective.