Rolling bearing degradation trend prediction method

Abstract

The invention relates to a rolling bearing degradation trend prediction method, wherein the method comprises the following steps: firstly, extracting characteristic parameters of a rolling bearing from multiple perspectives such as a time domain, a frequency domain and a time-frequency domain; then, fusing the multi-perspective features of vibration signals of the rolling bearing by using principal component analysis to construct a degradation trend curve; and finally, establishing a prediction model through the time convolutional neural network, predicting the degradation trend curve of the rolling bearing, and realizing accurate prediction of the degradation trend of the rolling bearing. According to the method, the degradation trend of the rolling bearing is predicted by adopting the time convolution network embedded with the dynamic convolution, the degradation state of the rolling bearing is reasonably evaluated, the multi-perspective features are fused by utilizing principal component analysis, the relevance information in a signal sequence is mined through the time convolution network, the feature extraction capability is improved, and the accuracy of the prediction model is improved. The degeneration critical state of the rolling bearing can be found in advance, and the degeneration trend can be accurately predicted.

Description

technical field [0001] The invention relates to the technical field of fault diagnosis based on deep learning, in particular to a method for predicting the degradation trend of rolling bearings. Background technique [0002] As a common rotating machinery in industrial production, rolling bearings are widely used in aerospace, intelligent manufacturing, wind power generation and other industrial fields. They play an irreplaceable role in maintaining the motion accuracy and work efficiency of rotating machinery. As one of the most vulnerable parts in industrial production, the faults of rotating equipment caused by rolling bearings account for about 30% of the total faults. [0003] In the existing technology, there are many methods for rolling bearing fault diagnosis. The research methods of rolling bearing degradation trend based on neural network algorithm mainly include recurrent neural network, deep belief network and multi-layer perceptron, etc. However, there are singl...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, there are many methods for rolling bearing fault diagnosis. The research methods of rolling bearing degradation trend based on neural network algorithm mainly include recurrent neural network, deep belief network and multi-layer perceptron, etc. However, there are single features and insufficient feature fusion. , weak feature extraction ability and other issues

Images