基于深度Q网络的城市轨道交通协同限流方法  

Cooperative Passenger Flow Control Method for Urban Rail Transit Utilizing Deep Q-Network

作　　者：王殿元 赵兴东 豆飞[2] 周旭 WANG Dianyuan;ZHAO Xingdong;DOU Fei;ZHOU Xu(Traffic Control Technology Co.,Ltd.,Beijing 100070;Beijing Subway Operation Co.,Ltd.,Beijing 100044)

机构地区：[1]交控科技股份有限公司,北京100070 [2]北京市地铁运营有限公司,北京100044

出　　处：《都市快轨交通》2024年第3期97-102,共6页Urban Rapid Rail Transit

基　　金：国家重点研发计划资助(2020YFB1600702);北京市科技新星计划项目(Z211100002121098)。

摘　　要：为解决城市轨道交通高峰小时区间满载率过高的问题,本文提出一种基于深度强化学习的城市轨道交通协同限流控制方法。该方法利用历史客流数据建立线网层面的限流仿真环境和智能体模型,以区间满载率为状态,以限流策略为动作,以客流体验为奖励,通过多轮强化学习训练产生最优的限流方案。随后利用北京地铁线网数据构建仿真实验并验证了该方法的有效性。仿真结果表明,协同限流方法可以有效降低断面客流量,缓解高峰小时区间拥挤程度,提高乘客出行舒适度。Rapid urbanization and population growth have led to a continuous increase in passenger flow in urban rail transit,which presents significant challenges to the safety,comfort,and stability of rail transit operations.To solve the problem of excessive load rate of urban rail transit during peak hours,we propose a cooperative passenger flow control method for urban rail transit based on deep reinforcement learning.This method uses the full load rate between intervals as its state,a flow restriction strategy as its action,and the passenger flow experience as its reward.It generates an optimal flow restriction scheme through multi-round reinforcement learning.We validated the effectiveness of this method by constructing simulation experiments using data from the Beijing subway network.The simulation results show that the cooperative passenger flow control method can effectively reduce passenger flow in a section,relieve congestion during peak hours,and improve passenger travel comfort.

关 键 词：城市轨道交通 深度强化学习 客流控制 北京地铁 

分 类 号：U231[交通运输工程—道路与铁道工程]