Mechanical part fault diagnosis method based on deep learning under data imbalance

Abstract

The invention discloses a mechanical part fault diagnosis method based on deep learning under data imbalance. The mechanical part fault diagnosis method comprises the steps: firstly obtaining an original vibration signal from a sensor, and obtaining frequency domain data through fast Fourier transform; then, inputting the frequency domain data into a generative adversarial network based on a Wasserstein distance; after multiple rounds of adversarial training of the generator and the discriminator, when the WGAN reaches Nash equilibrium, generating a large amount of fault sample data from the generator, and then mixing the generated fault sample data into original fault sample data to balance a data set; and finally, converting the balanced sample data into two-dimensional data, and inputting the two-dimensional data into a global average pooling convolutional neural network for feature extraction and fault classification to realize fault diagnosis of the mechanical parts. According tothe invention, the WGAN is used to reasonably solve the problem of data imbalance, and the GAPCNN is used to carry out fault classification diagnosis, so that the diagnosis precision is improved.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of mechanical equipment, and in particular relates to a fault diagnosis method for mechanical parts based on deep learning under data imbalance. Background technique [0002] With the advent of the Industry 4.0 era, the intelligent supervision of machinery and equipment has become the booster of the new generation of industrial revolution. Monitoring the operating status of mechanical parts by obtaining big data in service is the main focus of current research. However, in the actual production process, most of the operating data of mechanical parts collected by sensors are normal state data, and it is difficult to obtain fault state data, which leads to a serious imbalance in the characteristics of the collected data set. The diagnostic ability and generalization ability of the model generally decreased. How to obtain a complete data set with well-balanced characteristics in the mechanic...

AI Technical Summary

Problems solved by technology

The undersampling technique is the opposite of oversampling, that is, reducing the number of most samples so that the number is basically the same as that of a few samples. This artificial balance method lacks a theoretical basis, and the diagnostic results are still unsatisfactory.

In terms of algorithms, it is mainly to improve the traditional algorithm structure or design new algorithms. The mainstream methods include classifier integration methods, cost-sensitive methods, and feature selection methods. Although these methods try to avoid the impact of unbalanced data sets on diagnostic results, It has a certain effect in a specific field or a certain scene, but the effect is not obvious, and it cannot achieve the purpose of expanding data features. The trained model has low diagnostic accuracy and poor generalization ability.

Images