Residual network rolling bearing fault diagnosis method based on time-frequency analysis

Abstract

The invention relates to a residual network rolling bearing fault diagnosis method based on time-frequency analysis. The method comprises the following steps of S1. collecting vibration signal data, converting a vibration time domain signal of a rolling bearing into a time-frequency map by using short-time Fourier transform, and converting the time-frequency map into a two-dimensional gray level time-frequency map; and S2. performing feature extraction on the signal by using a residual network, and diagnosing a fault type of the bearing, wherein the input of the residual network is the gray level time-frequency map generated in the step S1, and the output of the residual network is a fault diagnosis result. According to the residual network rolling bearing fault diagnosis method provided by the invention, the bearing vibration data are converted into the time-frequency map by using the short-time Fourier transform, the time domain and frequency domain features during the vibration of afaulty bearing can be clearly reflected, and the network can accurately diagnose different fault types conveniently. Since the time-frequency signal contains both the time domain and frequency domaininformation of the bearing, and the deepening of a network layer of the residual network does not cause the problem of gradient disappearance or gradient explosion, higher accuracy can be obtained bythe method while performing fault diagnosis on the bearing.

Description

technical field [0001] The invention relates to the field of bearing fault diagnosis, in particular to a residual network rolling bearing fault diagnosis method based on time-frequency analysis. Background technique [0002] Rolling bearings are an important part of mechanical components. In large-scale equipment or production lines, there are a large number of rolling bearings in operation at any time. Once a serious fault occurs in the rolling bearing, it will make it difficult to control the accuracy of the product, and even stop the mechanical equipment or production line. Therefore, it is very important to carry out fault diagnosis on rolling bearings. When the bearing fails, it can be found and repaired in time, which is also of great help to the operation reliability of the equipment. [0003] Fault diagnosis of rolling bearings can use traditional methods, such as feature extraction of vibration signals, fault classification and other means. However, this method r...

AI Technical Summary

Problems solved by technology

[0005] Application number 201710747694.9, named as a rolling bearing fault diagnosis method based on convolutional neural network, using the short-time Fourier transform method to transform the rolling bearing from the time domain information to the frequency domain information, this method can extract the frequency domain characteristics of the signal , but this method uses an ordinary convolutional neural network in the subsequent processing. When the number of network layers deepens, the gradient disappears or the gradient explodes.

Therefore, the number of network layers is limited, the high-dimensional features of the bearing signal cannot be extracted, and the diagnostic accuracy is also limited

[0006] The application number is 201810339956.2, and the name is the establishment method of the rolling bearing intelligent diagnosis model based on the convolutional neural network. Although this method converts the one-dimensional signal of the bearing into a two-dimensional signal, the conversion method is arranged in order, so that the stacked two-dimensional The signal has no actual physical meaning and cannot reflect the time domain and frequency domain information of bearing faults

Therefore, the accuracy of bearing fault diagnosis cannot be improved.

[0007] Huang Chicheng's master's degree thesis of Zhejiang University, "Research on Rolling Bearing Fault Diagnosis Optimization Method Combining Time-Frequency Analysis and Convolutional Neural Network" used ResNet18 to diagnose the time-frequency signal of the bearing. This method is directly identified in the residual network. The network structure is not modified according to the characteristics of the bearing, and there is no visual analysis of the network training process

Images