A correlation-graph-CNN method for fault diagnosis of wind turbine based on state tracking and data driving model. (March 2023)

• Record Type: Journal Article
• Authors: Wang, Dongming; Cao, Chenyi; Chen, Naichao; Pan, Weiguo; Li, Hongchuan; Wang, Xudong
• Is Part Of: Sustainable energy technologies and assessments. Volume 56(2023)
• Publication Date: 2023-03
• Subjects: Wind turbine -- Fault diagnosis -- Supervisory control and data acquisition -- Correlation analysis -- Convolutional neural network; Renewable energy sources -- Periodicals; Energy development -- Technological innovations -- Periodicals; Electric power production -- Periodicals; Energy storage -- Periodicals; 333.79
• DOI: 10.1016/j.seta.2022.102995
• Languages: English
• ISSNs: 2213-1388

Abstract

Highlights: A correlation-graph-CNN method was proposed for fault diagnosis of wind turbine. State tracking was proposed to obtain more abundant information of fault. All the data were selected from SCADA system without adding any sensor. Pearson correlation analysis facilitates the increase in accuracy rate of the model. The accuracy rate of our proposed method can reach up to about 85%. Abstract: Due to the multidimensional parameters and the weak correlations with faults, it is still a challenge for the fault diagnosis of wind turbine based on the large amount of supervisory control and data acquisition (SCADA) data. In this work, a correlation-graph-convolutional neural network (CNN) method was proposed to develop a new methodology to predict the faults of wind turbine. We proposed a state tracking strategy by setting all the 24 h SCADA data before fault emergence as the input data. Meanwhile, the state parameters of wind turbine under normal status were also considered for comparison. Pearson correlation coefficient was conducted to quantitatively calculate the coupling level between state parameters, and a hotspot graph was designed to aggregate the correlation coefficients of all the state parameters to reconstruct an image that was adopted as the input data of CNN model. Finally, two practical faults were employed as the samples to study the performances of the correlation-graph-CNN method. The results showed that the accuracy of fault diagnosis can reach about 90%, that was greatly valuable in practical applications. Hence, our proposed method would have the important contribution on fault diagnosis of wind turbine.