Method and system for evaluating abrasion loss of railway track

By combining variational autoencoder with Gaussian process regression model, wavelet transform and local weighted regression processing, limit learning machine and principal component analysis, as well as Gaussian hybrid model and convolutional neural network methods, the insufficient data processing capability, accuracy and real-time performance in railway track wear evaluation is solved, and higher evaluation accuracy and adaptability are achieved.

CN119646631AActive Publication Date: 2025-03-18CHENGDU IND VOCATIONAL TECHN COLLEGE
3 Cites 0 Cited by

Patent Information

Application Number
CN202510153768.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-12
Publication Date
2025-03-18
Estimated Expiration
2045-02-12

AI Technical Summary

Technical Problem

The prior art lacks data processing capabilities, accuracy and real-time performance in the evaluation of railway track wear volume, making it difficult to meet the needs of complex railway operating environments.

Method used

Data cleaning and completion were performed by combining variational autoencoder and Gaussian process regression model, and the local characteristics of orbital wear data were extracted through wavelet transformation and local weighted regression processing, and data dimensionality reduction was performed by combining extreme learning machines and principal component analysis. Then, through Gaussian mixed model clustering processing, the prediction results are optimized by integrated convolutional neural network, and error adjustment is performed using adaptive weighted learning and gradient enhancement decision tree, and finally screen the evaluation results of high reliability through reliability analysis.

Benefits of technology

It significantly improves the accuracy, reliability and real-time performance of track wear evaluation, especially in multi-dimensional, diverse data processing and nonlinear relationship modeling, with higher prediction accuracy and stronger adaptability.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a railway track abrasion loss assessment method and system, and relates to the technical field of railway track abrasion loss assessment, and the method comprises the steps: obtaining train operation data and track historical abrasion data, and carrying out the data cleaning and complementing through a preprocessing model based on a variational auto-encoder and a Gaussian process regression model. Secondly, wavelet transformation and local weighted regression processing are carried out on the preprocessed data to extract local features and a linear relation, and a feature data set is constructed; dimensionality reduction is further carried out by combining an extreme learning machine with a principal component analysis model, and classification is carried out by using a Gaussian mixture model to generate a category feature set; and inputting the feature set into a prediction model, optimizing a prediction result in combination with a convolutional neural network, then carrying out reliability analysis, and taking a result of which the reliability is higher than a threshold value as a final track abrasion loss evaluation result. The method improves the accuracy and reliability of wear prediction, and is suitable for intelligent maintenance and management of railway tracks.
Need to check novelty before this filing date? Find Prior Art