Scholarlink

This paper addresses the issue of delays in high-speed rail trains, applying the Markov chain model to analyze the lateral propagation mechanism of delays and based on this, constructs an integrated CNN-LSTM-Attention model. The model incorporates the Maximum Information Coefficient-based MIC-BP feature selection algorithm to optimize the input feature set, effectively enhancing the model’s performance in practical applications. In the case study using Guangzhou-Shenzhen high-speed rail data, the model achieved a prediction accuracy of 97.71% within an allowable error margin of one minute, demonstrating its effectiveness in handling complex datasets, providing a basis for the adjustment of high-speed railway transportation organizations.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-Speed Train Delay Propagation and Prediction Based on Markov Chains and Ensemble Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xinqi Lu,
  • Pengju Shen,
  • Gang Zhou,
  • Liying Song,
  • Yingxu Chen

摘要

This paper addresses the issue of delays in high-speed rail trains, applying the Markov chain model to analyze the lateral propagation mechanism of delays and based on this, constructs an integrated CNN-LSTM-Attention model. The model incorporates the Maximum Information Coefficient-based MIC-BP feature selection algorithm to optimize the input feature set, effectively enhancing the model’s performance in practical applications. In the case study using Guangzhou-Shenzhen high-speed rail data, the model achieved a prediction accuracy of 97.71% within an allowable error margin of one minute, demonstrating its effectiveness in handling complex datasets, providing a basis for the adjustment of high-speed railway transportation organizations.