Optimal Energy Management Strategy for Extended-range Electric Vehicle via GA-BP Driving Pattern Recognition
-
摘要: 为了改善增程式电动汽车在复杂工况下的燃油经济性,提出一种利用遗传算法优化的反向传播神经网络(GA-BP)工况识别的规则能量管理策略。针对BP算法收敛速度慢、泛化能力差的问题,采用遗传算法优化BP神经网络。以中国工况中市区、市郊、高速作为工况类别,工况速度曲线为BP神经网络训练样本,构建遗传算法优化的BP神经网络识别器。在识别的工况下,以能耗费用为燃油经济性的评价指标,采用果蝇算法优化规则能量管理策略参数。仿真结果表明,使用遗传算法优化的BP神经网络识别正确率达到99.99%,优于未进行优化的神经网络识别;基于工况识别的能量管理策略,合理分配增程式电动汽车工作模式,有效的降低燃油消耗。Abstract: In order to improve the fuel economy of extended-range electric vehicle(E-REV) under complex driving condition, a rule energy management strategy based on BP neural network optimized by genetic algorithm is proposed. The genetic algorithm is used to optimize BP neural network for overcoming the shortcomings of slow convergence speed and poor generalization ability of BP neural network. City cycle of China include urban, suburban and expressway driving circle are used as the velocity curves of the three driving circles for training samples. Then, the rule energy management strategy is optimized under the identified driving pattern. The results show that the accuracy rate of driving pattern recognition of BP neural network optimized by genetic algorithm is 99.99%, which is higher than that without genetic algorithm. And the fuel economy of E-REV is improved under the energy management strategy based on driving pattern recognition, which is reasonable distribution of the working mode of the E-REV.
-
表 1 增程式电动汽车参数
参数单位 参数值 整车质量M/kg 6 000 轮胎半径r/m 0.376 0 发动机最高扭矩Tfc-max/(Nm) 250 发动机最高转速ωfc-max/(r·min-1) 6 000 发动机最高功率Pfc-max/kw 157 发电机最高扭矩Tgc-max/(Nm) 250 发电机最高转速ωgc-max/(r·min-1) 4 500 发电机最高功率Pgc-max/kW 117.8 驱动电机最高扭矩Tmc-max/(Nm) 850 驱动电机最高转速ωmc-max/(r·min-1) 4 000 驱动电机最高功率Pmc-max/kW 356 电池容量C/(A·h) 37 电池模块数n 16 主减速比i 6.143 表 2 典型工况及训练目标值
工况 目标值 归一化 市区 1 -1 市郊 2 0 高速公路 3 1 表 3 是否应用GA优化BP网络工况识别对比
识别方式 工况类型 最大误差 均方差 识别率 GA-BP 市区工况 0.0768 0.0173 100% 市郊工况 0.0185 0.0139 99.98% 高速公路工况 0.00000394 0.000295 100% 综合 0.0768 0.0033 99.99% BP 市区工况 0.6952 0.1615 60% 市郊工况 0.0615 0.0988 99.95% 高速公路工况 0.2272 0.0465 99.94% 综合 0.6952 0.1168 89.37 表 4 规则能量管理策略
驱动模式 工作状态 电池驱动纯电动模式 SOC>SOChigh、Pr < Pc-max 电池放电单独驱动 电池和增程器双动力混合驱动模式 SOClow≤SOC≤SOChigh、Plast≤Pr 发动机在发动机输出功率等于上一时刻输出功率 SOC>SOChigh、Pr≥Pc-max、P≤Pr 发动机处于最佳功率工点 增程器单独驱动且电池充电模式 SOC < SOClow 电池电量过低, 增程器输出的功率为充电功率与驱动需求功率之和 SOClow≤SOC≤SOChigh、Plast>Pr 发动机输出功率与上一秒时刻输出功率相等, 充电的功率为Plast-Pr SOC>SOChigh、Pr≥Pc-max、P>Pr 充电的功率为P-Pr 表 5 能量管理策略优化变量
优化变量 优化区间 SOClow 0.20~0.50 SOChigh 0.60~0.90 表 6 3种方式能耗费用
元 类型 油耗 电耗 总费用 未使用 23.24 -1.71 21.54 BP 24.39 -2.45 21.94 GA-BP 15.75 2.05 17.80 -
[1] GUANETTI J, FORMENTIN S, SAVARESI S M. Energy management system for an electric vehicle with a rental range extender: a least costly approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(11): 3022-3034 doi: 10.1109/TITS.2016.2529962 [2] LI J Q, JIN X J, XIONG R. Multi-objective optimization study of energy management strategy and economic analysis for a range-extended electric bus[J]. Applied Energy, 2017, 194: 798-807 doi: 10.1016/j.apenergy.2016.10.065 [3] 邓涛, 韩海硕, 罗俊林. 基于动态规划算法的混合动力汽车改进型ECMS能量管理控制研究[J]. 中国机械工程, 2018, 29(3): 326-332 doi: 10.3969/j.issn.1004-132X.2018.03.012DENG T, HAN H S, LUO J L. Improved ECMS energy management control of HEVs based on DP algorithm[J]. China Mechanical Engineering, 2018, 29(3): 326-332 (in Chinese) doi: 10.3969/j.issn.1004-132X.2018.03.012 [4] 邓涛, 罗俊林, 韩海硕, 等. 混合动力汽车工况识别自适应能量管理策略[J]. 西安交通大学学报, 2018, 52(1): 77-83 https://www.cnki.com.cn/Article/CJFDTOTAL-XAJT201801012.htmDENG T, LUO J L, HAN H S, et al. Adaptive energy management strategy based on driving cycle identification for hybrid electric vehicles[J]. Journal of Xi'an Jiaotong University, 2018, 52(1): 77-83 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XAJT201801012.htm [5] 熊演峰, 余强, 闫晟煜, 等. 基于GPS定位的插电式混合动力汽车能量管理策略研究[J]. 交通信息与安全, 2019, 37(6): 103-110 https://www.cnki.com.cn/Article/CJFDTOTAL-JTJS201906014.htmXIONG Y F, YU Q, YAN S Y, et al. An optimal control strategy of plug-in hybrid electric vehicle based on GPS position[J]. Journal of Transport Information and Safety, 2019, 37(6): 103-110 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JTJS201906014.htm [6] 刘永刚, 卢立来, 解庆波, 等. 基于道路坡度信息的插电式混合动力汽车能量管理策略[J]. 工程科学学报, 2016, 38(7): 1025-1031 https://www.cnki.com.cn/Article/CJFDTOTAL-BJKD201607019.htmLIU Y G, LU L L, XIE Q B, et al. Energy management for plug-in hybrid electric vehicles based on road gradient[J]. Chinese Journal of Engineering, 2016, 38(7): 1025-1031 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BJKD201607019.htm [7] 刘永刚, 解庆波, 秦大同, 等. 基于工况识别的混合动力汽车能量管理策略优化[J]. 机械传动, 2016, 40(5): 64-69, 73 https://www.cnki.com.cn/Article/CJFDTOTAL-JXCD201605015.htmLIU Y G, XIE Q B, QIN D T, et al. Energy management strategy optimization of HEV based on driving pattern recognition[J]. Journal of Mechanical Transmission, 2016, 40(5): 64-69, 73 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXCD201605015.htm [8] XIANG C L, DING F, WANG W D, et al. Energy management of a dual-mode power-split hybrid electric vehicle based on velocity prediction and nonlinear model predictive control[J]. Applied Energy, 2017, 189: 640-653 doi: 10.1016/j.apenergy.2016.12.056 [9] 方越栋. 基于随机动力需求预测的混合动力公交车能量管理策略[D]. 杭州: 浙江大学, 2015FANG Y D. Energy management strategies for hybrid electric bus based on stochastic power demand prediction[D]. Hangzhou: Zhejiang University, 2015 (in Chinese) [10] LI J, ZHOU Q, HE Y L, et al. Dual-loop online intelligent programming for driver-oriented predict energy management of plug-in hybrid electric vehicles[J]. Applied Energy, 2019, 253: 113617 doi: 10.1016/j.apenergy.2019.113617 [11] SUN C, HU X S, MOURA S J, et al. Velocity predictors for predictive energy management in hybrid electric vehicles[J]. IEEE Transactions on Control Systems Technology, 2015, 23(3): 1197-1204 doi: 10.1109/TCST.2014.2359176 [12] 秦大同, 彭志远, 刘永刚, 等. 基于工况识别的混合动力汽车动态能量管理策略[J]. 中国机械工程, 2014, 25(11): 1550-1555 doi: 10.3969/j.issn.1004-132X.2014.11.024QIN D T, PEN Z Y, LIU Y G, et al. Dynamic energy management strategy of HEV based on driving pattern recognition[J]. China Mechanical Engineering, 2014, 25(11): 1550-1555 (in Chinese) doi: 10.3969/j.issn.1004-132X.2014.11.024 [13] LEI Z Z, QIN D T, LIU Y G, et al. Dynamic energy management for a novel hybrid electric system based on driving pattern recognition[J]. Applied Mathematical Modelling, 2017, 45: 940-954 doi: 10.1016/j.apm.2017.01.036 [14] WANG J, HUANG Y, XIE H M, et al. Driving pattern recognition and energy management for extended range electric bus[C]//Proceedings of 2014 IEEE Vehicle Power and Propulsion Conference. Coimbra: IEEE, 2014: 1-6 [15] RIZZO F, CARACOGLIA L. Artificial neural network model to predict the flutter velocity of suspension bridges[J]. Computers & Structures, 2020, 233: 106236 [16] 俞美鑫, 施卫, 蒋龙, 等. 基于GA-BP神经网络的动力锂电池SOC估算[J]. 电子技术应用, 2020, 46(1): 104-107, 112 https://www.cnki.com.cn/Article/CJFDTOTAL-DZJY202001031.htmYU M X, SHI W, JIANG L, et al. SOC estimation of power lithium battery based on GA-BP neural network[J]. Application of Electronic Technique, 2020, 46(1): 104-107, 112 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DZJY202001031.htm [17] ZOU M, XUE L, GAI H J, et al. Identification of the shear parameters for lunar regolith based on a GA-BP neural network[J]. Journal of Terramechanics, 2020, 89: 21-29 doi: 10.1016/j.jterra.2020.02.003 [18] DESHWAL D, SANGWAN P, KUMAR D. A language identification system using hybrid features and back-propagation neural network[J]. Applied Acoustics, 2020, 164: 107289 doi: 10.1016/j.apacoust.2020.107289 [19] 王嵘冰, 徐红艳, 李波, 等. BP神经网络隐含层节点数确定方法研究[J]. 计算机技术与发展, 2018, 28(4): 31-35 doi: 10.3969/j.issn.1673-629X.2018.04.007WANG R B, XU H Y, LI B, et al. Research on method of determining hidden layer nodes in BP neural network[J]. Computer Technology and Development, 2018, 28(4): 31-35 (in Chinese) doi: 10.3969/j.issn.1673-629X.2018.04.007 [20] 郑立平, 郝忠孝. 遗传算法理论综述[J]. 计算机工程与应用, 2003, 39(21): 50-53, 96 doi: 10.3321/j.issn:1002-8331.2003.21.017ZHEN L P, HAO Z X. A review on the theory for the genetic algorithm[J]. Computer Engineering and Applications, 2003, 39(21): 50-53, 96 (in Chinese) doi: 10.3321/j.issn:1002-8331.2003.21.017 [21] 李敏强, 寇纪淞, 林丹. 等. 遗传算法的基本理论与应用[M]. 北京: 科学出版社, 2003LI M Q, KOU J S, LIN D, et al. Basic theory and application of genetic algorithm[M]. Beijing: Science Press, 2003 (in Chinese) [22] DING G S, DONG F Z, ZOU H F, et al. Fruit fly optimiza-tion algorithm based on a hybrid adaptive-cooperative learning and its application in multilevel image thresholding[J]. Applied Soft Computing, 2019, 84: 105704 doi: 10.1016/j.asoc.2019.105704