考虑司机行为的城市轨道交通列车运行速度曲线优化
曹江1万欣2张程熙2付敏雪2郝世聪2
Optimization of Urban Rail Transit Train Operation Speed Curves Considering Driver Behavior
CAO JiangWAN XinZHANG ChengxiFU MinxueHAO Shicong
-
作者信息:1.中车青岛四方机车车辆股份有限公司， 266111， 青岛
2.北京交通大学电子信息工程学院， 100044， 北京
-
Affiliation:CRRC Qingdao Sifang Locomotive & Rolling Stock Co., Ltd., 266111, Qingdao, China
-
关键词:
-
Key words:
-
DOI:10.16037/j.1007-869x.2023.10.016
-
中图分类号/CLCN:U284.48
-
栏目/Col:研究报告
摘要:
目的：鉴于目前对于城市轨道交通人工驾驶场景下辅助司机完成驾驶的需求考虑不足，为了进一步结合人工驾驶场景对列车运行曲线进行优化，需要对考虑司机行为的列车运行曲线优化方法进行研究。方法：基于人工驾驶场景的特点和实际列车操纵方法，构建了列车运行速度曲线优化模型。对司机行为习惯进行了调研，并制定了相应的生成驾驶建议的原则。采用AGMOPSO(自适应网格多目标粒子群优化)算法对列车运行曲线进行了优化，并介绍了一种生成驾驶建议的方法。对京广铁路保定东站—石家庄站的列车实际运行数据进行了仿真计算；引入UMD(统一司机行为)模型构建了人处于闭环条件下的仿真环境，并进行了仿真分析。结果及结论：京广铁路保定东站—石家庄站的列车实际运行数据仿真计算，验证了生成驾驶建议方法的有效性。人在闭环仿真表明，列车运行速度曲线优化后，列车运行能耗为1 82458 kWh，列车运行时间为2 049 s，满足准时和节能的目标，并符合司机的行为习惯。
Abstracts:
Objective: In view of the current inadequate consideration of assisting drivers in completing their driving needs against urban rail transit manual operation scenarios, further optimization of train operation curves with the integration of manual operation scenarios is required. It is aimed to investigate methods for optimizing train operation curves considering driver behavior.Method: A train operation speed curve optimization model is established based on the characteristics of manual operation scenarios and actual train operation methods. Driver behavior patterns are surveyed, and corresponding principles for generating driving behavior recommendations are formulated. The AGMOPSO (adaptive grid multiobjective particle swarm optimization) algorithm is applied to optimize train operation curves, and a method for generating driving behavior recommendations is introduced. Simulation calculations are performed using the actual train operation data from Baoding East Station to Shijiazhuang Station on the BeijingGuangzhou Railway. The UMD (unified driver behavior) model is incorporated to create a closedloop simulation environment for manual operation, followed by an analysis of human behavior in this closedloop simulation.Result & Conclusion: Simulation calculations based on the actual train operation data from Baoding East Station to Shijiazhuang Station on the BeijingGuangzhou Railway validate the effectiveness of the driving behavior recommendation generation method. Closedloop simulation with human behavior analysis reveals that, following the optimization of train operation speed curves, the train consumes 1 824.58 kWh of energy, operates for 2049 seconds, meeting the goals of punctuality and energy efficiency and conforming to driver behavior patterns.
- 上一篇: 高铁物流运输模式及其可行性
- 下一篇: TEST-20231219
过刊浏览
资料下载
友情链接