Visual image-based permanent magnet driving motor demagnetization fault diagnosis model construction method and fault diagnosis method and system

Abstract

The invention discloses a visual image-based permanent magnet driving motor demagnetization fault diagnosis model construction method and a fault diagnosis method and system, and the method comprises the steps: enabling a motor surface magnetic leakage signal to serve as an original signal for fault diagnosis, converting the original signal into a two-dimensional Fourier spectrogram, and obtaining a two-dimensional Fourier spectrogram; and global features and local features are extracted and fused. And finally, constructing a classifier by using the fused features, and particularly, carrying out fault diagnosis by using a preferable softmax classifier. According to the invention, two-dimensional Fourier transform is adopted to convert a one-dimensional time domain signal into a spectrogram, and effective fault features hidden in the time domain signal are displayed; the visual features of the image are extracted through an auto-encoder method, the effectiveness of the features is improved, the softmax classifier which is simple in structure and small in calculation amount is adopted for fault diagnosis, and the problem of demagnetization fault diagnosis of the permanent magnet driving motor for the electric vehicle is effectively solved.

Description

technical field [0001] The invention relates to a motor demagnetization fault diagnosis technology, in particular to a visual image-based permanent magnet drive motor demagnetization fault diagnosis model construction method and a fault diagnosis method and system, which can be especially applied to electric vehicles. Background technique [0002] my country's new energy vehicle industry has gone through 20 years of development. my country has become the world's largest producer and consumer of new energy vehicles. Among them, the scale of pure electric vehicles accounts for more than 50% of the world, ranking first in the world. International advanced. With the continuous improvement of the intelligence and integration of new energy vehicles, their internal structures are becoming increasingly complex. In the past two years, the safety accidents of new energy vehicles have been increasing, and the industry has gradually changed from "mileage anxiety" to "safety anxiety". Sa...

AI Technical Summary

Problems solved by technology

[0006] 1) Traditional feature extraction based on one-dimensional time domain or frequency domain signals requires complex signal preprocessing and strong professional requirements

[0007] 2) Global feature extraction cannot be applied to image occlusion, rotation, etc., and local features cannot take into account the low efficiency of features caused by comprehensive features

Images