Bearing fault diagnosis method based on semi-supervised adversarial network

Abstract

The invention discloses a bearing fault diagnosis method based on a semi-supervised adversarial network. The method comprises the following steps: S100, collecting a vibration signal xf when a bearing has a true fault, a vibration signal xh when the bearing normally operates, and a vibration signal of a to-be-detected bearing; S200, constructing a semi-supervised generative adversarial network composed of a generator g, a feature network f, a fault classifier fc, a discriminator d, an auxiliary classifier ac and a diagnosis network diag, and training the semi-supervised generative adversarial network: S201, training the generator g to generate a fault state and a pseudo bearing vibration signal under normal operation; S202, training the feature network f, the fault classifier fc, the discriminator d and the auxiliary classifier ac according to the vibration signals xf and xh and the pseudo bearing vibration signal; and S203, after training convergence of the step S201 and the step S202, training the diagnosis network diag by using the vibration signals xh and xf and the pseudo bearing vibration signal; and S300, inputting the vibration signal of the to-be-detected bearing into the trained diagnosis network diag for fault diagnosis.

Description

technical field [0001] The disclosure belongs to the technical field of bearing fault diagnosis, and in particular relates to a bearing fault diagnosis method based on a semi-supervised confrontation network. Background technique [0002] For high-end equipment such as high-speed trains and aerospace vehicles, each failure means huge property losses and irreparable casualties, and bearing condition monitoring and fault diagnosis are one of the core tasks of its fault prediction and health management (PHM) First, the deep learning method represented by the convolutional neural network has achieved great success in the bearing fault diagnosis task. This achievement is due to its network structure characteristics and training methods, such as local weight sharing. However, in the actual application process, the running status of bearings of high-end equipment such as high-speed trains, wind power equipment, and aero-engines can be monitored in real time through sensors, so as t...

AI Technical Summary

Problems solved by technology

For example, based on the support vector data description machine learning method to build a binary tree model for mechanical fault diagnosis, however, the construction of the binary tree is limited by the degree of imbalance of different fault categories

[0004] In summary, the existing bearing fault diagnosis methods for data category imbalance, including data-based methods and classifier-based methods, can alleviate the negative impact of sample imbalance to a certain extent, but in practical applications However, there are still some problems: 1) The requirement for the number of samples is high, and it is difficult to adapt to the diagnostic task under extremely unbalanced conditions; 2) The randomness of sample generation is strong, and it is difficult to achieve reliable category-specific directional enhancement; 3) The model cannot be fully utilized during the learning process. 4) Issues such as information integrity of hidden variables other than fault category labels are not considered in sample generation

Images