An eco-driving algorithm for trains through distributing energy: A Q-Learning approach

被引：55
作者：
Zhu, Qingyang [1 ,2 ]
Su, Shuai [1 ,2 ]
Tang, Tao [1 ]
Liu, Wentao [1 ]
Zhang, Zixuan [1 ]
Tian, Qinghao [3 ]
机构：
[1] Beijing Jiaotong Univ, State Key Lab Rail Traff Control & Safety, Beijing 100044, Peoples R China
[2] China Acad Railway Sci Corp Ltd, Ctr Natl Railway Intelligent Transportat Syst Eng, Beijing 100081, Peoples R China
[3] Beijing Jiaotong Univ, Sch Sci, Beijing 100044, Peoples R China
来源：
ISA TRANSACTIONS | 2022年 / 122卷 / 24-37期
基金：
中国国家自然科学基金; 北京市自然科学基金;
关键词：
Driving strategy; Eco-driving; Q-Learning; OPERATION; OPTIMIZATION; SUBWAY; SYSTEM;
D O I：
10.1016/j.isatra.2021.04.036
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
The energy-efficient train operation methodology is the focus of this paper, and a Q-Learningbased eco-driving approach is proposed. Firstly, the core idea of energy-distribution-based method (EDBM) that converts the eco-driving problem to the finite Markov decision process is presented. Secondly, Q-Learning approach is proposed to determine the optimal energy distribution policy. Specifically, two different state definitions, i.e., trip-time-relevant (TT) and energy-distribution-relevant (ED) state definitions, are introduced. Finally, the effectiveness of the proposed approach is verified in a deterministic and a stochastic environment. It is also illustrated that TT-state approach takes about 20 times more computation time compared with ED-state approach while the space complexity of TTstate approach is nearly constant. The hyperparameter sensitivity analysis demonstrates the robustness of the proposed approach. (C) 2021 ISA. Published by Elsevier Ltd. All rights reserved.
引用
收藏
页码：24 / 37
页数：14
相关论文
共 43 条
  • [1] The key principles of optimal train control-Part 2: Existence of an optimal strategy, the local energy minimization principle, uniqueness, computational techniques
    Albrecht, Amie
    Howlett, Phil
    Pudney, Peter
    Vu, Xuan
    Zhou, Peng
    [J]. TRANSPORTATION RESEARCH PART B-METHODOLOGICAL, 2016, 94 : 509 - 538
  • [2] The key principles of optimal train control-Part 1: Formulation of the model, strategies of optimal type, evolutionary lines, location of optimal switching points
    Albrecht, Arnie
    Howlett, Phil
    Pudney, Peter
    Vu, Xuan
    Zhou, Peng
    [J]. TRANSPORTATION RESEARCH PART B-METHODOLOGICAL, 2016, 94 : 482 - 508
  • [3] Aradi S, 2013, INT SYMP COMP INTELL, P135, DOI 10.1109/CINTI.2013.6705179
  • [4] SOLUTION OF THE PROBLEM OF THE ENERGETICALLY OPTIMAL-CONTROL OF THE MOTION OF A TRAIN BY THE MAXIMUM PRINCIPLE
    ASNIS, IA
    DMITRUK, AV
    OSMOLOVSKII, NP
    [J]. USSR COMPUTATIONAL MATHEMATICS AND MATHEMATICAL PHYSICS, 1985, 25 (06): : 37 - 44
  • [5] Brenna M., 2016, INT J VEH TECHNOL, V2016, P1
  • [6] An Energy Efficient Train Dispatch and Control Integrated Method in Urban Rail Transit
    Bu, Bing
    Qin, Guoying
    Li, Ling
    Li, Guojie
    [J]. ENERGIES, 2018, 11 (05)
  • [7] Application of fuzzy predictive control technology in automatic train operation
    Cao, Yuan
    Ma, Lianchuan
    Zhang, Yuzhuo
    [J]. CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS, 2019, 22 (Suppl 6): : 14135 - 14144
  • [8] Research on dynamic nonlinear input prediction of fault diagnosis based on fractional differential operator equation in high-speed train control system
    Cao, Yuan
    Zhang, Yuzhuo
    Wen, Tao
    Li, Peng
    [J]. CHAOS, 2019, 29 (01)
  • [9] Multiobjective Overtaking Maneuver Planning for Autonomous Ground Vehicles
    Chai, Runqi
    Tsourdos, Antonios
    Al Savvaris
    Chai, Senchun
    Xia, Yuanqing
    Chen, C. L. Philip
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2021, 51 (08) : 4035 - 4049
  • [10] Chen Rong-wu, 2011, Application Research of Computers, V28, P2126, DOI 10.3969/j.issn.1001-3695.2011.06.034
12345