Scholarlink

<p>As an integral component of urban public transportation systems, urban rail transit significantly alleviates urban road traffic pressure and holds a pivotal position in the development strategies of urban public transportation. This paper aims to investigate the energy-saving optimization control problem of trains under the influence of events. Firstly, we consider train operation time, train dynamics, and train energy consumption and then establish an energy-saving control optimization model for rail transit that takes into account the influence of events. Secondly, we devise a deep reinforcement learning-based method for optimizing train energy-saving control, which incorporates the impact of events. Finally, we validate our proposed method using real-world urban rail transit data in a subway simulation environment, comparing it with other deep reinforcement learning methods. The results indicate that the proposed approach effectively reduces energy consumption for rail transit trains when facing both unforeseen and regular events, across different subway systems.</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

Optimization of energy-efficient control in rail transit systems under event impact

  • 期刊论文

  • 发表日期：

    2025年6月10日

查看全文
  • Changxi Ma,
  • Mingxi Zhao,
  • Tao Wang

摘要

As an integral component of urban public transportation systems, urban rail transit significantly alleviates urban road traffic pressure and holds a pivotal position in the development strategies of urban public transportation. This paper aims to investigate the energy-saving optimization control problem of trains under the influence of events. Firstly, we consider train operation time, train dynamics, and train energy consumption and then establish an energy-saving control optimization model for rail transit that takes into account the influence of events. Secondly, we devise a deep reinforcement learning-based method for optimizing train energy-saving control, which incorporates the impact of events. Finally, we validate our proposed method using real-world urban rail transit data in a subway simulation environment, comparing it with other deep reinforcement learning methods. The results indicate that the proposed approach effectively reduces energy consumption for rail transit trains when facing both unforeseen and regular events, across different subway systems.