基于KL散度工况识别的混合动力汽车队列的分层控制

doi:10.3969/j.issn.1674-8484.2024.02.013

摘要/Abstract

摘要：

针对混合动力汽车队列行驶过程中工况的适应性问题，提出了一种基于KL（Kullback- Leibler）散度工况识别的分层控制方法。上层控制器利用车—车通信技术，获取队列中前车状态信息，采用模型预测控制（MPC）算法，实现队列纵向控制，并计算出最优跟车车速；下层控制器基于典型工况，离线求解需求功率的转移概率矩阵，并通过Q-Learning算法训练最优Q表嵌入整车模型中；在行驶中以固定时间长度在线更新转移概率矩阵，采用KL散度进行工况识别，根据识别的工况类型，结合当前时刻车速、需求功率和电池荷电状态（SOC），通过在线查表实现转矩分配。结果表明：与未考虑工况识别策略相比，本策略的油耗降低了8.6%；与作为基准的动态规划（DP）相比，增加了4.8%；在与DP油耗基本保持相同的前提下，该策略离线仿真时间缩短21%，不仅能够在线应用，还能实时适应工况变化。

关键词: 混合动力汽车, 汽车队列, 工况识别, 模型预测控制（MPC）算法, Q-Learning算法, KL（Kullback- Leibler）散度

Abstract:

A hierarchical control method was proposed for adapting the working condition in the driving process for hybrid-electric-vehicle (HEV) platoon based on a Kullback-Leibler (KL) divergence working condition recognition algorithm. The upper layer controller utilized vehicle-vehicle communication technology, obtained state information of the leading vehicle in the platoon. It adopted the Model Predictive Control (MPC) algorithm, achieved longitudinal control of the platoon, and calculated the optimal speed for the following vehicle. The lower layer controller, initially based on typical working conditions, solved the transition probability matrix of demanded power offline. Trained the optimal Q-table offline, embedded it in the complete vehicle model by the Q-Learning algorithm. During driving, the transition probability matrix was updated online at regular intervals, and KL divergence was used to recognition the working conditions. According to the identified working conditions types, combined the current moment vehicle speed, the demanded power, and the battery state of charge (SOC), the torque allocation were achieved through an online lookup table. The results show that the fuel consumption of this strategy is reduced by 8.6% compared with the strategy without considering the condition identification, and increased by 4.8% compared with the dynamic programming (DP) as the benchmark. Under the premise of maintaining the same fuel consumption as DP, the off-line simulation time is reduced by 21%, which can not only be applied online, but also adapt to the change of working conditions in real time.

Key words: hybrid electric vehicle (HEV), vehicle platoon, working condition recognition, model predictive control (MPC) algorithm, Q-Learning algorithm, KL (Kullback-Leibler) divergence

中图分类号:

U461.8

尹燕莉, 王福振, 詹森, 黄学江, 张鑫新, 张富椿. 基于KL散度工况识别的混合动力汽车队列的分层控制[J]. 汽车安全与节能学报, 2024, 15(2): 242-252.

YIN Yanli, WANG Fuzhen, ZHAN Sen, HUANG Xuejiang, ZHANG Xinxin, ZHANG Fuchun. Hierarchical control for hybrid electric vehicle platoon based on KL divergence working condition recognition[J]. Journal of Automotive Safety and Energy, 2024, 15(2): 242-252.

图/表 21

图1 并联式HEV结构

整备质量，m	1.372 t
迎风面积，A	2.0 m²
车轮半径，R	0.272 m
主减速器传动比，i₀	4.38
传动比，i_g	0.685~3.425
发动机最大功率，P_e,max	63 kW
发动机最大转矩，T_e,max	110 Nm
电动机最大功率，P_m,max	10 kW
电动机最大转矩，T_e,max	46.5 Nm
电池容量，Q	6.5 Ah

表1 并联式HEV主要参数

图2 发动机油耗MAP图

图3 电机效率模型

图4 等效电池模型

图5 PF通信拓扑结构

图6 3类典型工况

图7 需求功率转移概率矩阵分布图

图8 WLTC工况下不同车速离线优化转矩MAP图

图9 CLTC-P工况

图10 CLTC-P工况识别结果

图11 基于KL工况识别的HEV队列分层控制流程

图12 CLTC-P工况下队列中各车车速曲线

图13 CLTC-P工况下队列中各车位移曲线

图14 CLTC-P工况下队列中各车加速度曲线

图15 在线识别结果

图16 CLTC-P工况下发动机输出转矩

图17 CLTC-P工况下电动机输出转矩

图18 CLTC-P工况下电池SOC的变化

	100 km燃油消耗体积 / L
	KL-QL	QL	DP
领航车	4.072	4.415	3.874
跟随车1	4.219	4.550	4.025
跟随车2	4.085	4.430	3.915
平均油耗	4.125	4.465	3.938
比较 / %	+4.8	+13.4	—

表2 CLTC-P工况下的100 km燃油消耗

	仿真时间 / s		相对减少率
	QL	DP	相对减少率
离线	4 680	5 631	-21%
在线	396	—	—

表3 CLTC-P工况下的仿真时间

参考文献 19

[1]	孙超, 刘波, 孙逢春. 新能源汽车节能规划与控制技术研究综述[J]. 汽车安全与节能学报, 2022, 13(4): 593-616.
	SUN Chao, LIU Bo, SUN Fengchun. Review of energy-saving planning and control technology for new energy vehicles[J]. *J Auto Safety Energy*, 2022, 13(4): 593-616. (in Chinese)
[2]	ZAHNG Fengqi, HU Xiaosong, Langari R, et al. Energy management strategies of connected HEVs and PHEVs: Recent progress and outlook[J]. *Prog Energy Combu Sci*, 2019, 73: 235-256. doi: 10.1016/j.pecs.2019.04.002 URL
[3]	胡云峰, 刘迪, 赵靖华, 等. 智能网联环境下车辆能耗与排放优化控制的研究现状与展望[J]. 中国公路学报, 2022, 35(3): 1-14. doi: 10.19721/j.cnki.1001-7372.2022.03.001
	HU Yunfeng, LIU Di, ZHAO Jinghua, et al. Research status and prospect of vehicle energy consumption and emission optimization control in intelligent network environment[J]. *China J Highway Transp*, 2022, 35(3): 1-14. (in Chinese)
[4]	鲁若宇, 胡杰, 陈瑞楠, 等. 基于DMPC的智能汽车协同式自适应巡航控制[J]. 汽车工程, 2021, 43(8): 1177-1186.
	LU Ruoyu, HU Jie, CHEN Ruinan, et al. Cooperative adaptive cruise control of intelligent vehicles based on DMPC[J]. *Auto Engineering*, 2021, 43(8): 1177-1186. (in Chinese)
[5]	罗捷, 鲁良叶, 何德峰, 等. 通信拓扑切换下车辆队列分布式模型预测控制[J]. 控制理论与应用, 2021, 38(7): 887-896.
	LUO Jie, LU Liangye, HE Defeng, et al. Distributed model predictive control of vehicle platoons with switching communication topologies[J]. *Contr Theo Appl*, 2021, 38(7): 887-896. (in Chinese)
[6]	张毅, 裴华鑫, 姚丹亚. 车路协同环境下车辆群体协同决策研究综述[J]. 交通运输工程学报, 2022, 22(3): 1-18.
	ZHANG Yi, PEI Huaxin, YAO Danya. Research review on cooperative decision-making for vehicle swarms in a vehicle-infrastructure cooperative environment[J]. *J Transp Eng*, 2022, 22(3): 1-18. (in Chinese)
[7]	Ye Xin, LAI Fei, HUO Zhiwei. Energy management strategy design and simulation validation of hybrid electric vehicle driving in an intelligent fleet[J]. *Electronics*, 2019, 8(12): 1516. doi: 10.3390/electronics8121516 URL
[8]	洪金龙, 高炳钊, 董世营, 等. 智能网联汽车节能优化关键问题与研究进展[J]. 中国公路学报, 2021, 34(11): 306-334. doi: 10.19721/j.cnki.1001-7372.2021.11.025
	HONG Jinlong, GAO Bingzhao, DONG Shiying, et al. Key problems and research progress of energy saving optimization for intelligent connected vehicles[J]. *Chin J Highway Transp*, 2021, 34(11): 306-334. (in Chinese)
[9]	尹燕莉, 黄学江, 潘小亮, 等. 基于PID与Q-Learning的混合动力汽车队列分层控制[J]. 吉林大学学报(工学版), 2023, 53(5): 1481-1489.
	YIN Yanli, HUANG Xuejiang, PAN Xiaoliang, et al. Hierarchical control of hybrid electric vehicle platooning based on PID and Q-Learning algorithm[J]. *J Jilin Univ (Eng Tech Ed): 2023*, 53(5): 1481-1489. (in Chinese)
[10]	HU Xiaosong, ZHANG Xiaoqian, TANG Xiaolin, et al. Model predictive control of hybrid electric vehicles for fuel economy, emission reductions, and inter-vehicle safety in car-following scenarios[J]. *Energy*, 2020, 196: 117101. doi: 10.1016/j.energy.2020.117101 URL
[11]	钱立军, 陈健, 吴冰, 等. 考虑驾驶员操作误差的混合动力汽车队列生态驾驶控制[J]. 汽车工程, 2021, 43(07): 1037-1045. doi: 10.19562/j.chinasae.qcgc.2021.07.011
	QIAN LiJun, CHEN Jian, WU Bing, et al. Eco-driving control for hybrid electric vehicle platoon consideration driver operation error[J]. *Auto Engineering*, 2021, 43(07): 1037-1045. (in Chinese)
[12]	YIN Yanli, HUANG Xuejiang, ZHAN Sen, et al. Hierarchical model predictive control strategy based on Q-Learning algorithm for hybrid electric vehicle platoon[J]. **Proc Inst Mech Eng, Part D: J Auto Eng**, 2022: 09544070221130826.
[13]	TANG Xiaolin, DUAN Ziwen, HU Xiaosong, et al. Improving ride comfort and fuel economy of connected hybrid electric vehicles based on traffic signals and real road information[J]. *IEEE Trans Veh Tech*, 2021, 70(4): 3101-3112. doi: 10.1109/TVT.2021.3063020 URL
[14]	WANG Yichun, ZHANG Yuanzhi, ZHANG Caizhi, et al. Genetic algorithm-based fuzzy optimization of energy management strategy for fuel cell vehicles considering driving cycles recognition[J]. *Energy*, 2023, 263: 126112. doi: 10.1016/j.energy.2022.126112 URL
[15]	连静, 常静, 李琳辉, 等. 基于模糊在线识别的并联混合动力客车自适应控制策略[J]. 北京理工大学学报, 2016, 36(3): 264-270.
	LIAN Jing, CHANG Jing, LI Linhui, et al. Adaptive control strategy based on online fuzzy recognition for parallel hybrid electric bus[J]. *J Beijing Univ Tech*, 2016, 36(3): 264-270. (in Chinese)
[16]	SHI Qin, QIU Duoyang, HE Lin, et al. Support vector machine-based driving cycle recognition for dynamic equivalent fuel consumption minimization strategy with hybrid electric vehicle[J]. *Adva Mech Engi*, 2018, 10(11): 1-17.
[17]	刘灵芝, 张冰战, 蒋通. 基于工况自适应的PHEV等效燃油最小策略[J]. 汽车安全与节能学报, 2020, 11(3): 371-378.
	LIU Lingzhi, ZHANG Bingzhang, JIANG Tong. Equivalent consumption minimization strategy for PHEV based on driving condition adaptation[J]. *J Auto Safe Energy*, 2020, 11(3): 371-378. (in Chinese)
[18]	LIU Qifang, DONG Shiying, YANG Zheng, et al. Energy management strategy of hybrid electric vehicles based on driving condition prediction[J]. *IFAC-Papers Online*, 2021, 54(10): 265-270.
[19]	YIN Yanli, HUANG Xuejiang, PAN Xiaoliang, et al. Hyperparameters of Q-Learning algorithm adapting to the driving cycle based on KL driving cycle recognition[J]. *Int’l J Auto Tech*, 2022, 23(4): 967-981.