A Method for Motor Fault Diagnosis Based on Graph Attention Network

Abstract

A motor fault diagnosis method based on a graph attention network, comprising the following steps: 1), dividing data samples; 2), performing band-pass filtering on the two-phase currents in each small sample to obtain denoised state characteristic voltages signal; 3), constructing a graph network: constructing a current graph network according to the extreme points of the state characteristic electrical signal, and then merging n current graph networks of two-phase currents in a large sample to obtain a two-phase current graph network of the large sample; 4 ), building a classification model based on graph attention networks: constructing graph attention networks based on two-phase current graph networks, and then fusing the features extracted by the two networks for classification. The electrical signal data used in the present invention is more convenient and low in cost in terms of data collection, and proposes a method of converting the signal into a graph network, which can effectively analyze the Motor faults are classified and diagnosed.

Description

technical field [0001] The invention relates to a motor fault diagnosis method based on a graph attention network, and belongs to the field of motor fault diagnosis. Background technique [0002] Motor bearings are one of the most critical components in a motor. Any bearing failure, even a small one, can lead to the failure of the entire system. According to statistics, 40-70% of electromechanical drive system and motor failures are caused by damage to rolling bearings. Therefore, the detection and diagnosis of motor faults is very important. Early detection of motor faults can prevent the bearing from accidentally causing the system to shut down, and ensure the continuous operation of the system while ensuring safety, thereby stabilizing the work efficiency. [0003] At present, the motor vibration signal can convey the health status information of the motor when it is running, so the mainstream motor fault diagnosis methods are mainly based on this signal to analyze the ...

AI Technical Summary

Problems solved by technology

[0004] Traditional signal processing and feature extraction techniques, such as Fourier transform, wavelet decomposition, empirical mode decomposition, and Silbert transform, are not ideal when applied to motor fault diagnosis based on motor current signals

Images