Residual life prediction method for rolling bearing based on DL Gamma

DOI:	10.11809/bqzbgcxb2023.09.020
Keywords:	residual life; rolling bearing; deep learning; performance degradation; gamma process
Abstract:	Aiming at the difficulty of feature extraction and life distribution of rolling bearing, a life prediction method of rolling bearing based on DL Gamma is proposed by combining Deep Learning and Gamma Process. The typical time domain and frequency domain characteristics of the vibration signal and the time frequency diagram of the original signal obtained by the continuous wavelet transform are used as one dimensional time series data and two dimensional image sequence data, respectively. The hybrid input network is constructed by combining the deep convolution neural network and the bidirectional long short term memory network. The one dimensional time series data and the two dimensional image sequence data are input into the network to obtain the performance degradation factor. The gamma process is used to model the performance degradation factor curve to realize the residual life prediction. The experimental results show that the method can effectively predict the residual life of rolling bearings, and the prediction results overcome the shortcomings of the traditional method, which is beneficial to practical application.
Issue:	Vol. 44 No. 9 (2023)
Published:	2023-09-28
PDF HTML