Scholarlink

<p>Accurate and rapid anomaly detection of train control systems is an inevitable requirement for ensuring the safe and efficient operation of high-speed railways. Currently, the manual offline fault diagnosis has issues such as ineffectiveness in fault locating and a relatively large scope of fault impact. In response, an anomaly detection model based on multimodal learning with the attention mechanism is proposed. According to the interrelated relationship between text logs and visual images representing equipment working status, a language-vision fusion two-stream multimodal neural network learning architecture is designed. The entire network structure, is centered around the attention mechanism. At the first stage, independent text and visual processing streams are introduced, focusing separately on the context of the text and the changes in indicator light display at specific positions on the visual images, learning the heterogeneity of each modality. At the later stage, in the decision-making layer, the learning results of language and vision are organically fused through logical operations, culminating in the output of anomalous states for the train control system. Causation experiment and comparative tests on real train operation datasets demonstrate the model’s superior performance in terms of precision of 98.63% and recall of 98.82% compared to other methods, validating its effectiveness.</p>

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

Multimodal anomaly detection for high-speed train control system based on attention mechanism

  • 期刊论文

  • 发表日期：

    2026年2月27日

查看全文
  • Renwei Kang,
  • Yanzhi Pang,
  • Jianfeng Cheng,
  • Lei Liu,
  • Jianqiu Chen

摘要

Accurate and rapid anomaly detection of train control systems is an inevitable requirement for ensuring the safe and efficient operation of high-speed railways. Currently, the manual offline fault diagnosis has issues such as ineffectiveness in fault locating and a relatively large scope of fault impact. In response, an anomaly detection model based on multimodal learning with the attention mechanism is proposed. According to the interrelated relationship between text logs and visual images representing equipment working status, a language-vision fusion two-stream multimodal neural network learning architecture is designed. The entire network structure, is centered around the attention mechanism. At the first stage, independent text and visual processing streams are introduced, focusing separately on the context of the text and the changes in indicator light display at specific positions on the visual images, learning the heterogeneity of each modality. At the later stage, in the decision-making layer, the learning results of language and vision are organically fused through logical operations, culminating in the output of anomalous states for the train control system. Causation experiment and comparative tests on real train operation datasets demonstrate the model’s superior performance in terms of precision of 98.63% and recall of 98.82% compared to other methods, validating its effectiveness.