A novel method of combining generalized frequency response function and convolutional neural network for complex system fault diagnosis

• Published in: PloS one Vol. 15; no. 2; p. e0228324
• Main Authors: Chen, Lerui, Zhang, Zerui, Cao, Jianfu
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 04.02.2020; Public Library of Science (PLoS)
• Subjects: Accuracy; Adaptive algorithms; Algorithms; Artificial neural networks; Biology and Life Sciences; Cable television broadcasting industry; Classification; Complex systems; Computer and Information Sciences; Convolution; Engineering and Technology; Fault diagnosis; Feature extraction; Frequency dependence; Frequency response functions; Information systems; Laboratories; Learning; Manufacturing; Medical diagnosis; Medicine and Health Sciences; Methods; Monitoring systems; Neural networks; Neural Networks, Computer; Novels; Permanent magnets; Physical Sciences; Principal Component Analysis; Research and Analysis Methods; Spectrum analysis; Support Vector Machine; Synchronous motors; Wavelet transforms; United States; China
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0228324

To solve the problem of low accuracy in traditional fault diagnosis methods, a novel method of combining generalized frequency response function(GFRF) and convolutional neural network(CNN) is proposed. In order to accurately characterize system state information, this paper proposed a variable step size least mean square (VSSLMS) adaptive algorithm to calculate the second-order GFRF spectrum values under normal and fault states; In order to improve the ability of fault feature extraction, a convolution neural network (CNN) with gradient descent learning rate and alternate convolution layer and pooling layer is designed to extract the fault features from GFRF spectrum. In the proposed method, the second-order GFRF spectrum of each state of Permanent Magnet Synchronous Motor (PMSM) is obtained by VSSLMS; Then, the two-dimension GFRF spectrum, which is regarded as the gray value of the image,will be further transformed into image. Finally, the CNN is trained with learning rate by gradient descent way to realize the fault diagnosis of PMSM. Experimental results indicate that the accuracy of proposed method is 98.75%, which verifies the reliability of the proposed method in application of PMSM fault diagnosis.