铁道运输与经济2024，Vol.46Issue(12)：169-177,9.DOI:10.16668/j.cnki.issn.1003-1421.2024.12.18
面向新型列控系统的高速列车智能驾驶策略优化方法
Intelligent Driving Strategy Optimization Method for High Speed Trains with New Train Control System
摘要
Abstract
The new train control system has the function of moving blocks,which can further improve the transportation efficiency of high speed railways.The existing C2/C3+ATO technology cannot be directly applied to the train tracking operation scenario under the moving block condition.To address this issue,the research on the optimization of intelligent driving strategy for high speed trains oriented to the new train control system was carried out.Under the reinforcement learning framework,the problem was transformed into a Markov decision problem.The state set,action set,and reward function of the train tracking operation were designed,and the static and dynamic constraints of the train tracking operation were taken into account.A multi-train tracking operation environment was constructed based on the train dynamics model,and a deep Q-network-based intelligent train driving strategy optimization method was developed.Three simulation cases were designed to verify the effectiveness of the proposed method.The simulation results show that the deep Q-network algorithm can quickly obtain the driving strategy curve of the tracking train compared with the genetic algorithm,and the following train can realize energy-saving and safe tracking operation under the premise of punctuality.The research can provide a reference for the optimization of intelligent driving strategies for high speed trains.关键词
新型列控系统/列车智能驾驶/移动闭塞/深度Q网络/强化学习Key words
New Train Control System/Intelligent Train Driving/Moving Block/Deep Q-Network/Reinforcement Learning分类
交通工程引用本文复制引用
侯卓璞,刘斌,刘雅晴,李一楠..面向新型列控系统的高速列车智能驾驶策略优化方法[J].铁道运输与经济,2024,46(12):169-177,9.基金项目
中国铁道科学研究院集团有限公司科研项目(2021YJ018) （2021YJ018）
北京华铁信息技术有限公司科研项目(2023HT06) （2023HT06）
