Bearing Fault Diagnosis Based on BP Neural Network and Transfer Learning

Rolling bearing is widely used in rotating machinery, which is one of the most prone to failure of industrial parts. In order to obtain the running state of rolling bearing timely and accurately, this paper uses Fourier transform to preprocess the data; based on the transfer component analysis to narrow the difference between the labeled data and the unlabeled data distribution, it is convenient to extract the feature information; BP neural network algorithm is used to build the network model, and then the data is tested, so as to realize the detection of bearing fault state And diagnosis. The experimental results show that the transfer learning based on principal component analysis can calibrate the features of unlabeled data well, and BP neural network can identify the fault types of rolling bearing well.

[1]  Yisheng Zou,et al.  A novel transfer learning method for bearing fault diagnosis under different working conditions , 2021 .

[2]  Mingxuan Liang,et al.  Rolling bearing fault diagnosis based on feature fusion with parallel convolutional neural network , 2020, The International Journal of Advanced Manufacturing Technology.

[3]  Zhiming Wang,et al.  Bearing Intelligent Fault Diagnosis Based on Wavelet Transform and Convolutional Neural Network , 2020, Shock and Vibration.

[4]  Shaojiang Dong,et al.  The fault diagnosis method of rolling bearing under variable working conditions based on deep transfer learning , 2020, Journal of the Brazilian Society of Mechanical Sciences and Engineering.

[5]  Omair Inam,et al.  Transfer learning in deep neural network based under-sampled MR image reconstruction. , 2020, Magnetic resonance imaging.

[6]  Ran Tao,et al.  Optimized sparse fractional Fourier transform: Principle and performance analysis , 2020, Signal Process..

[7]  Hee-Jun Kang,et al.  Rolling element bearing fault diagnosis using convolutional neural network and vibration image , 2019, Cognitive Systems Research.

[8]  Weiping Zhang,et al.  Multi-parameter online measurement IoT system based on BP neural network algorithm , 2018, Neural Computing and Applications.

[9]  Jianhua Cai,et al.  Bearing fault diagnosis method based on the generalized S transform time–frequency spectrum de-noised by singular value decomposition , 2018, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science.

[10]  Yunzhou Zhang,et al.  Joint transfer component analysis and metric learning for person re‐identification , 2018, Electronics Letters.

[11]  Aini Hussain,et al.  Optimal BP neural network algorithm for state of charge estimation of lithium-ion battery using PSO with PCA feature selection , 2017 .