Energy Management of Connected and Automated Hybrid Electric Vehicles based on Reinforcement Learning
编号:76 访问权限:仅限参会人 更新:2021-12-03 10:13:24 浏览:136次 张贴报告

报告开始:暂无开始时间(Asia/Shanghai)

报告时间:暂无持续时间

所在会场:[暂无会议] [暂无会议段]

暂无文件

摘要
As one of the most important parts of intelligent transportation system(ITS), vehicles are developing in the direction of intellectualization and electrification. At present, the research on connected and automated vehicles focuses on its safety and trafficability, and lacks the exploration of its economy. However, energy conservation is also one of the important goals of ITS. In this study, an energy management method of connected and automated hybrid electric vehicles is proposed, which takes into account the trafficability and safety as well as the economy. Based on Dijkstra algorithm, the length of the road is weighted according to the real-time traffic data to ensure the trafficability of the planning path.In addition, according to the obtained path parameters and vehicle dynamics model constraints, the safety velocity of connected and automated hybrid electric vehicles is planned. Besides, based on the planning velocity, the method of reinforcement learning is used to control the working mode and energy output of the engine and motors to achieve energy saving. The results show that using the proposed energy management method in this study, the connected and automated hybrid electric vehicles not only can improve the traffic efficiency, but also achieve the effect of energy saving.
关键词
CICTP
报告人
Shengguang Xiong
Wuhan University of Technology

稿件作者
Shengguang Xiong Wuhan University of Technology
发表评论
验证码 看不清楚,更换一张
全部评论
重要日期
  • 会议日期

    12月17日

    2021

    12月20日

    2021

  • 12月16日 2021

    报告提交截止日期

  • 12月24日 2021

    注册截止日期

主办单位
Chinese Overseas Transportation Association
Chang'an University
联系方式
移动端
在手机上打开
小程序
打开微信小程序
客服
扫码或点此咨询