A rail safety performance evaluation method and platform integrating multiple parameters

By using the hierarchical analysis method and ELM network method in the evaluation of orbital safety performance, the evaluation inaccuracy caused by parameter deviation in the prior art is solved, and higher evaluation accuracy and effectiveness are achieved.

CN118569647BActive Publication Date: 2025-03-28XIAN JIAOTONG ENG COLLEGE
2 Cites 0 Cited by

Patent Information

Application Number
CN202410846247.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2025-03-28
Estimated Expiration
2044-07-12

AI Technical Summary

Technical Problem

When evaluating track safety performance, the prior art is difficult to effectively deal with parameters that deviate from normal data, resulting in inaccurate evaluation results and increasing the frequency of inspection and maintenance of the transportation department.

Method used

A fusion multi-parameter track safety performance evaluation method is adopted to collect parameters that affect track performance from each sensor, establish a parameter matrix, and analyze the impact matrix using hierarchical analysis method. Then, the adaptive coefficient is determined using the ELM network to form an adaptive ELM network, and finally the feature vector is input to the network to obtain the track safety performance evaluation value.

Benefits of technology

The accuracy of track safety performance evaluation is improved, making the characteristic channel parameters learned by the model more in line with the actual situation, reducing the difference values ​​of non-important data points, and enhancing the effectiveness of the evaluation.

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

Abstract

The present invention provides a track safety performance evaluation method and platform integrating multiple parameters, collecting influencing parameters that affect the track performance safety to be evaluated; establishing a parameter matrix of similar influencing parameters in time sequence, and using the hierarchical analysis method to analyze the parameter matrix to obtain an influence matrix of similar influencing parameters; then determining the adaptive coefficient of each channel in the trained ELM network; forming a feature vector with the influencing parameters at the current moment; and inputting the feature vector into the adaptive ELM network model to obtain the track safety performance evaluation value. In the present invention, the adaptive coefficient of each channel is obtained by calculating the influencing parameter, and in the process of obtaining, the difference value generated by non-important data is weakened by searching the search space, and then the adaptive coefficient is solved by using the adaptive function, so that the learning content of the ELM network model changes according to the input object, and the track safety performance can be evaluated more accurately, thereby improving the safety of railway operation.
Need to check novelty before this filing date? Find Prior Art