Traction motor health diagnosis method and system

By combining equal-angle resampling and signal preprocessing with signal time-frequency analysis, and integrating an improved gradient boosting tree diagnostic model based on gray wolf packs, the problems of signal non-stationarity and noise interference in traction motor fault diagnosis using electrical signal analysis methods are solved, enabling accurate extraction and diagnosis of fault characteristic indicators.

CN115329810BActive Publication Date: 2026-07-03ZHUZHOU CSR TIMES ELECTRIC CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ZHUZHOU CSR TIMES ELECTRIC CO LTD
Filing Date
2022-08-08
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Existing electrical signal analysis methods are difficult to accurately extract fault feature indicators in traction motor fault diagnosis. They are affected by strong background noise such as torque fluctuations and power supply harmonics, and the non-stationarity of the signal makes it difficult to extract and identify fault feature indicators.

Method used

A signal preprocessing method combining equal-angle resampling, variational mode decomposition, and Wiener filtering is adopted to eliminate the influence of speed, operating conditions, and load. Fault characteristic indicators are extracted by combining signal time-frequency analysis, and an improved gradient boosting tree diagnostic model based on gray wolf packs is applied for fault diagnosis.

Benefits of technology

It effectively eliminates diagnostic errors caused by power supply harmonics, variable speed, and variable load, improves the accuracy of non-stationary weak signal feature extraction and fault diagnosis, and realizes accurate identification and severity assessment of traction motor fault modes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115329810B_ABST
    Figure CN115329810B_ABST
Patent Text Reader

Abstract

The application provides a traction motor health diagnosis method and system, the method comprising: obtaining a plurality of existing electrical signals of a traction motor from a traction control unit of a converter, and performing equal-angle resampling and data preprocessing on the plurality of existing electrical signals respectively to obtain a plurality of equal-angle dimensionless stationary signals irrelevant to rotation speed, working condition and load; performing demodulation on the plurality of equal-angle dimensionless stationary signals respectively and extracting a plurality of fault feature indexes by using a signal time-frequency analysis method; taking the plurality of fault feature indexes as inputs, applying a preset multiple-input multiple-output diagnosis model to perform fault diagnosis, and outputting a fault mode of the traction motor and a severity corresponding to the fault mode; and performing health management on the traction motor according to the plurality of fault feature indexes. The application can effectively eliminate diagnosis errors caused by PWM power supply harmonics, variable rotation speed and variable load transient processes, and improve the accuracy of non-stationary weak signal feature extraction and fault diagnosis.
Need to check novelty before this filing date? Find Prior Art