Sliding bearing fault diagnosis method based on generative adversarial network and convolutional neural network

Abstract

The invention provides a sliding bearing fault diagnosis method based on a generative adversarial network and a convolutional neural network. The method comprises the steps of: collecting vibration signals and bearing bush temperatures of a sliding bearing under no fault and different fault conditions, and carrying out preprocessing; carrying out ensemble average empirical mode decomposition on the vibration signals to obtain an optimal intrinsic mode function and a characteristic frequency corresponding to the optimal intrinsic mode function; constructing state matrixes on the basis of the bearing bush temperatures, and respectively inputting the state matrixes into the generative adversarial network according to categories to generate samples; carrying out edge expansion on the data of an initial sample set, and setting fault state labels; constructing a sliding bearing convolutional neural network model, and training the model by using the data set and the corresponding fault labels; and collecting data of the current sliding bearing, constructing a state matrix, inputting the state matrix into the trained neural network model, and carrying out fault diagnosis and prediction. With the method of the invention adopted, the problems of unobvious vibration fault signals, poor diagnosis effect, insufficient bearing bush temperature utilization and insufficient sample size are effectively solved.

Description

technical field [0001] The invention relates to the technical field of fault diagnosis, in particular to a sliding bearing fault diagnosis method based on generating confrontation networks and convolutional neural networks. Background technique [0002] In the industrial manufacturing process, sliding bearings are an important component of mechanical components, and many fault states of rotating machinery are related to sliding bearings. Sliding bearings are also one of the links most prone to damage in the manufacturing process. Common faults include: excessive bearing clearance, oil film whirl, oil film oscillation, bearing wear, bush burning, bodiless, etc. These faults will reduce the rotation accuracy of the bearing, generate noise, vibration, and further cause the bearing to be stuck, and finally make the entire rotating machine fail. Therefore, it is of great significance to carry out fault diagnosis on sliding bearings. [0003] The traditional sliding bearing faul...

AI Technical Summary

Problems solved by technology

And when the temperature of the bearing bush exceeds the warning value, there are many reasons for the temperature rise of the bearing bush, and it is impossible to make an accurate diagnosis

At present, the popular deep learning-based fault diagnosis method combines feature extraction and pattern classification into one architecture, but it has high requirements for the number of samples.

And when dealing with high-latitude vibration signals, it takes a lot of manual feature extraction

Increased workload and difficulty in troubleshooting

Images