基于XGBoost模型的地铁列车运行状态仿真方法
门元昊1吴亮2刘晓双1秦枭喃1罗森林1
A Metro Train Running State Simulation Method Based on XGBoost Model
MEN YuanhaoWU LiangLIU XiaoshuangQIN XiaonanLUO Senlin
-
作者信息:1.北京理工大学信息系统及安全对抗实验中心， 100081， 北京
2.通号城市轨道交通技术有限公司， 100081， 北京
-
Affiliation:Information System and Security & Countermeasures Experimental Center, Beijing Institute of Technology, 100081, Beijing, China
-
关键词:
-
Key words:
-
DOI:10.16037/j.1007-869x.2022.03.022
-
中图分类号/CLCN:U29-39
-
栏目/Col:研究报告
摘要:
基于物理模型及列车性能参数的地铁列车运行状态仿真方法存在着列车适配性差、参数调整成本高等问题。为了给ATC(列车自动控制)系统的研究提供更为准确、高效的模拟试验平台，提出一种基于XGBoost(极端梯度提升)算法的列车运行状态仿真模型构建方法。该方法可从实际运行数据中学习列车的控制与运行特性，可针对不同的线路环境、不同列车车型实现更准确的列车运行状态仿真。在某地铁线路采用实车进行试验，结果表明：该方法建立的列车运行状态仿真模型准确、有效，可满足实际应用需求。
Abstracts:
The simulation method of metro train running state based on physical model and train performance parameters has the problems of poor train adaptability and high cost of parameter adjustment. In order to provide a more accurate and efficient simulation test platform for the research of ATC (automatic train control) system, a method for constructing simulation model of train running state based on XGBoost (extreme gradient boosting) is proposed. This method can acquire control and running characteristics of the train from actual operation data, and achieve more accurate running state simulation for different line environments and different types of train. Real vehicle is launched on certain metro line for experiments. The results show that the simulation model of train running state established by this method is accurate and effective, and meets the needs of practical application.
引文 / Ref:
门元昊，吴亮，刘晓双，等.基于XGBoost模型的地铁列车运行状态仿真方法[J].城市轨道交通研究，2022，25（3）：102.
MEN Yuanhao，WU Liang，LIU Xiaoshuang，et al.A Metro Train Running State Simulation Method Based on XGBoost Model[J].Urban mass transit，2022，25（3）：102.
MEN Yuanhao，WU Liang，LIU Xiaoshuang，et al.A Metro Train Running State Simulation Method Based on XGBoost Model[J].Urban mass transit，2022，25（3）：102.
- 上一篇: 高铁物流运输模式及其可行性
- 下一篇: 发展高速磁浮助力交通强国
过刊浏览
资料下载
友情链接