摘要
为提高插电式混合动力汽车的燃油经济性,对基于动态规划与机器学习的能量管理算法展开了研究。利用K-均值聚类算法将20个标准工况划分为3个类型的工况段,利用动态规划(DP)算法最优功率分配数据分别训练3个类型工况段的神经网络模型,在控制过程中根据实际工况段类型选择相应的神经网络模型进行功率分配,并对上海市某个随机工况进行了仿真运算,结果表明,该算法燃油经济性较电量消耗-电量维持(CD-CS)策略有明显的改善。
In order to improve the fuel economy of plug-in hybrid electric vehicle,this paper studies energy management algorithm based on dynamic programming and machine learning.Firstly,the K-means clustering algorithm is used to divide 20 standard driving cycles into three types of driving conditions.Secondly,neural network model of three driving condition types is trained with optimum power allocation of DP algorithm.And the corresponding neural network model is selected according to the actual driving condition type for power distribution.Finally,the simulation is carried out based on a random driving condition in Shanghai.The results show that the fuel economy of this algorithm is significantly improved compared with the CD-CS strategy.
作者
陈渠
殷承良
张建龙
秦文刚
Chen Qu;Yin Chengliang;Zhang Jianlong;Qin Wengang(National Engineering Laboratory for Automotive Electronic Control Technology,Shanghai Jiao Tong University,Shanghai 200240;United Automotive Electronic Systems Co.,Ltd.,Shanghai 201206)
出处
《汽车技术》
CSCD
北大核心
2020年第10期51-57,共7页
Automobile Technology
基金
上海汽车工业科技发展基金会项目(1745)。
