Method and device for identifying deterioration category of traction load harmonics

By verifying the validity of traction load harmonic degradation data, a training dataset was constructed and a deep learning model was used, which solved the problem of difficulty in judging the validity of input data and improved the accuracy of harmonic degradation type identification and training efficiency.

CN115169392BActive Publication Date: 2026-06-09CHINA ACADEMY OF RAILWAY SCI CORP LTD +2
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
CHINA ACADEMY OF RAILWAY SCI CORP LTD
Filing Date
2022-06-28
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

In existing technologies, the validity of traction load harmonic degradation input data used to train deep learning models is difficult to determine in advance, resulting in wasted time and resources during the training process.

Method used

By acquiring and verifying the validity of the original data on harmonic degradation of traction loads, a training dataset was constructed. A deep learning model was then used for training to obtain a harmonic degradation type identification model. Based on the voltage and current data characteristics of the harmonic degradation type identification model, data with low correlation were eliminated to improve the model accuracy.

Benefits of technology

This improved the accuracy of harmonic degradation type identification, reduced the consumption of manpower and material resources, and increased the efficiency of model training.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115169392B_ABST
    Figure CN115169392B_ABST
Patent Text Reader

Abstract

The application discloses a traction load harmonic deterioration category identification method and device, and relates to the field of electrical fault diagnosis of railway locomotives and vehicles. The method comprises the following steps: obtaining traction load harmonic deterioration data of a motor train unit; and obtaining the traction load harmonic deterioration category according to the traction load harmonic deterioration data and a pre-created harmonic deterioration category identification model, wherein the harmonic deterioration category identification model is pre-trained by traction load harmonic deterioration training data that has passed effectiveness verification. Before training the harmonic deterioration category identification model, the application pre-verified the effectiveness of the traction load harmonic deterioration original data used for training the model, ensured the effectiveness of the training data, and thus improved the accuracy of harmonic category identification using the harmonic deterioration identification model and reduced the consumption of manpower and resources.
Need to check novelty before this filing date? Find Prior Art