EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhanced intelligent train operation algorithms for metro train based on expert system and deep reinforcement learning

Yunhu Huang iD, Wenzhu Lai, Dewang Chen, Geng Lin, Jiateng Yin

PLOS One · 2025

DOI: 10.1371/journal.pone.0323478 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Yunhu Huang und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Data Science, Minjiang University, Fuzhou, Fujian, China. Hauptthemen laut MEDLINE: Deep Learning, Algorithms, Expert Systems, Railroads. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fujian Provinces Education Research Project for Young and Middle-aged Teachers, Minjiang University Talent Introduction Research Project, Innovation Star Talent Program of the Third Batch in Fujian Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In recent decades, automatic train operation (ATO) systems have been gradually adopted by many metro systems, primarily due to their cost-effectiveness and practicality. However, a critical examination reveals computational constraints, adaptability to unforeseen conditions and multi-objective balancing that our research aims to address. In this paper, expert knowledge is combined with deep reinforcement learning algorithm (Proximal Policy Optimization, PPO) and two enhanced intelligent train operation algorithms (EITO) are proposed. The first algorithm, EITO E , is based on an expert system containing expert rules and a heuristic expert inference method. On the basis of EITO E , we propose EITO P algorithm using the PPO algorithm to optimize multiple objectives by designing reinforcement learning strategies, rewards, and value functions. We also develop the double minimal-time distribution (DMTD) calculation method in the EITO implementation to achieve longer coasting distances and further optimize the energy consumption. Compared with previous works, EITO enables the control of continuous train operation without reference to offline speed profiles and optimizes several key performance indicators online. Finally, we conducted comparative tests of the manual driving, intelligent driving algorithm (ITOR, STON), and the algorithms proposed in this paper, EITO, using real line data from the Yizhuang Line of Beijing Metro (YLBS). The test results show that the EITO outperform the current intelligent driving algorithms and manual driving in terms of energy consumption and passengers’ comfort. In addition, we further validated the robustness of EITO by selecting some complex lines with speed limits, gradients and different running times for testing on the YLBS. Overall, the EITO P algorithm has the best performance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningAlgorithmsExpert SystemsRailroadsHumansReinforcement, PsychologyReinforcement Machine Learning

Autor:innen und Institutionen

  • College of Computer and Data Science, Minjiang University, Fuzhou, Fujian, China.
  • Digital Banking Laboratory of Industrial and Commercial Bank of China Software Development Center, Zhuhai, Guangdong, China.
  • School of Transportation, Fujian University of Technology, Fuzhou, Fujian, China.
  • Fujian BeiDou Navigation and Intelligent Transportation Collaborative Innovation Center, Fuzhou, Fujian, China.
  • State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China.

ORCID-Profile: Yunhu Huang

Förderung

  • Fujian Provinces Education Research Project for Young and Middle-aged Teachers
  • Minjiang University Talent Introduction Research Project
  • Innovation Star Talent Program of the Third Batch in Fujian Province
  • Special Fund for Education and Scientific Research of Fujian Provincial Department of Finance
  • Scientific Research Foundation of Fujian University of Technology
  • The Natural Science Foundation of Fujian Province
  • Science and Education Joint Special Project of Minjiang University
  • 2023 National College Students' Innovation and Entrepreneurship Training Program

Zitieren

APA

Huang, Y., Lai, W., Chen, D., Lin, G., & Yin, J. (2025). Enhanced intelligent train operation algorithms for metro train based on expert system and deep reinforcement learning. PLOS One. https://doi.org/10.1371/journal.pone.0323478

Vancouver

Huang Y, Lai W, Chen D, Lin G, Yin J. Enhanced intelligent train operation algorithms for metro train based on expert system and deep reinforcement learning. PLOS One. 2025. doi:10.1371/journal.pone.0323478

BibTeX
@article{Huang2025,
  author = {Huang, Yunhu and Lai, Wenzhu and Chen, Dewang and Lin, Geng and Yin, Jiateng},
  title = {Enhanced intelligent train operation algorithms for metro train based on expert system and deep reinforcement learning},
  journal = {PLOS One},
  year = {2025},
  doi = {10.1371/journal.pone.0323478}
}
Kostenlos starten →