Wind turbine fault diagnosis based on ReliefF-PCA and DNN

作者：

Highlights：

• A hybrid fault diagnosis method is proposed based on ReliefF, PCA and DNN.

• ReliefF algorithm is used to extract fault sensitive features.

• PCA algorithm is used for dimensionality reduction.

• The optimized ReliefF-PCA-DNN model increases the fault-diagnosis accuracy of the WTs.

摘要

•A hybrid fault diagnosis method is proposed based on ReliefF, PCA and DNN.•ReliefF algorithm is used to extract fault sensitive features.•PCA algorithm is used for dimensionality reduction.•The optimized ReliefF-PCA-DNN model increases the fault-diagnosis accuracy of the WTs.

论文关键词：Wind turbine,Fault diagnosis,ReliefF,Principal component analysis,Deep neural network

论文评审过程：Received 9 October 2020, Revised 1 March 2021, Accepted 7 April 2021, Available online 13 April 2021, Version of Record 22 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115016