强化学习在混合动力汽车能量管理上的研究综述

doi:10.16638/j.cnki.1671-7988.2021.016.049

汽车实用技术 ›› 2021, Vol. 46 ›› Issue (16): 178-180.DOI: 10.16638/j.cnki.1671-7988.2021.016.049

强化学习在混合动力汽车能量管理上的研究综述

赵春领

重庆交通大学机电与车辆工程学院

出版日期:2021-08-30 发布日期:2021-08-30
通讯作者: 赵春领
作者简介:赵春领（1997—），男，车辆工程硕士，重庆交通大学机电与车辆工程学院，主要从事混合动力汽车传动与控制方面的研究。
基金资助:
基金项目：重庆交通大学研究生科研创新项目（CYS20288）；重庆市技术创新与应用发展专项-重大主题专项项目（Z2311200003）。

Review of Research on Reinforcement Learning in Hybrid Electric Vehicle Energy Management

CHEN Xin, HOU Hongling

College of Mechatronics & Vehicle Engineering, Chongqing Jiaotong University

Online:2021-08-30 Published:2021-08-30
Contact: ZHAO Chunling

摘要/Abstract

摘要： 混合动力汽车被认为是平衡长久里程和低能耗的可行性技术途径，而能量管理策略技术是混合动力汽车的关键技术之一，目前强化学习（RL）在混合动力汽车能量管理策略开始得到了应用，其是制定有效能量管理策略的重要方法。文章对混合动力汽车能量管理问题进行了描述并分析总结了 RL 算法以及基于 RL 的混合算法在混合动力汽车能量管理上的各种应用现状以及趋势。

关键词: 混合动力汽车；RL 算法；能量管理

Abstract: Hybrid electric vehicles are considered to be a feasible technical approach to balance long-term mileage and low energy consumption, and energy management strategy technology is one of the key technologies of hybrid electric vehicles. At present, reinforcement learning (RL) is beginning to be applied in hybrid electric vehicle energy management strategies, which is an important method for formulating effective energy management strategies. In this article, the energy manage -ment problems of hybrid electric vehicles are described, analyzed and summarized, and various application status and trends of RL algorithms and RL-based hybrid algorithms in the energy management of hybrid electric vehicles are summarized.

Key words: Hybrid electric vehicle; Reinforcement learning algorithm; Energy management

赵春领. 强化学习在混合动力汽车能量管理上的研究综述[J]. 汽车实用技术, 2021, 46(16): 178-180.

CHEN Xin. Review of Research on Reinforcement Learning in Hybrid Electric Vehicle Energy Management[J]. Automobile Applied Technology, 2021, 46(16): 178-180.

参考文献

[1] 赵秀春,郭戈.混合动力电动汽车能量管理策略研究综述[J].自动化学报,2016,042(003):321-334. [2] 张冰战.插电式混合动力电动汽车能量管理策略研究[D]:合肥:合肥工业大学,2011. [3] 解少博,陈欢,刘通,等.基于DP-ECMS的插电式混合动力城市客车能量管理策略研究[J].汽车工程, 2017(07):12-17+57. [4] Liu T,Hu X,Hu W,et al.A heuristic planning reinforcement learningbased energy management for power-split plug-in hybrid electric vehicles[J].IEEE Transactions on Industrial Informatics, 2019,15 (12):6436-6445. [5] 隗寒冰,贺少川.基于深度 RL 的插电式柴电混合动力汽车多目标优化控制策略[J].重庆交通大学学报(自然科学版),2021,40(01): 44-52. [6] 蔡伟,何太碧.基于工况预测的混合动力汽车能量管理策略研究综述[J].时代农机,2020,47(01):15+18. [7] 刘腾.混合动力车辆 RL 能量管理研究[D].北京:北京理工大学, 2017. [8] 李茹杨,彭慧民,李仁刚,等.强化学习算法与应用综述[J].计算机系统应用, 2020,29(12):17-29. [9] Fang Y, Song C, Xia B, et al. An energy management strategy for hybrid electric bus based on reinforcement learning[C]//2015 27th Chinese Control and Decision Conference (CCDC). IEEE, 2015. [10] 陶吉利,谢亮,马龙华,等.一种基于 Q-learning 和规则的混合动力车辆运行实时能源管理方法, 201811468385.9 [P]. 2019-04-19. [11] Liu C, Yi L M. Power management for Plug-in Hybrid Electric Vehicles using Reinforcement Learning with trip information[C]// 2014 IEEE Transportation Electrification Conference and Expo (ITEC). IEEE, 2014. [12] Xu B, Hu X, Tang X, et al. Ensemble Reinforcement Learning -Based Supervisory Control of Hybrid Electric Vehicle for Fuel Economy Improvement[J]. IEEE Transactions on Transportation Electrification, 2020, 6(2):717-727. [13] N Yang, Han L, Xiang C, et al. Energy management for a hybrid electric vehicle based on blended reinforcement learning with backward focusing and prioritized sweeping[J]. IEEE Transactions on Vehicular Technology, 2021, PP(99):1. [14] 张昊,范钦灏,王巍,等.基于 RL 的多燃烧模式混合动力能量管理策略[J].汽车工程,2021,43(5): 683-691. [15] 隗寒冰,贺少川.基于深度RL的插电式柴电混合动力汽车多目标优化控制策略[J].重庆交通大学学报(自然科学版),2021,40(01): 44-52. [16] 王勇,何洪文,彭剑坤,等. 2020.基于深度RL的插电式混合动力汽车能量管理[C]//中国汽车工程学会年会暨展览会.上海；中国汽车工程学会：10-27. [17] Zhou J, Xue S, Xue Y, et al. A novel energy management strategy of hybrid electric vehicle via an improved TD3 deep reinforcement learning[J].Energy, 2021(5):118-120. [18] Du G, Y Zou, Zhang X, et al. Deep reinforcement learning based energy management for a hybrid electric vehicle[J]. Energy, 2020 (201):117-591. [19] 周涛,陆惠玲,霍兵强.深度信念网络研究进展[J].计算机工程与应用,2020,952(09):30-38. [20] 邵俊捷,陈国锋,于闯,等.基于 LSTM 循环神经网络的转向架部件监测参数趋势预测[J].计算机科学与应用, 2021, 11(6):8.

强化学习在混合动力汽车能量管理上的研究综述

Review of Research on Reinforcement Learning in Hybrid Electric Vehicle Energy Management

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics