依据智能交通的PHEV自适应指数模型预测控制  

PHEV Adaptive Exponent Model Predictive Control Depending on Intelligent Transportation System

作　　者：赵亮[1] 胡丰宾 ZHAO Liang;HU Feng-bin(Shanxi Institute of Mechanical and Electrical Engineering,Shanxi Changzhi 046011,China;Beiqi Futian Automobile Co.,Ltd.Beijing 101100,China)

机构地区：[1]山西机电职业技术学院,山西长治046011 [2]北汽福田汽车股份有限公司,北京101100

出　　处：《机械设计与制造》2021年第10期145-149,共5页Machinery Design & Manufacture

基　　金：山西省教育科学“十三五”规划2018年度规划课题(GH-18151)。

摘　　要：为了降低插电式混合动力汽车使用成本,提出了基于智能交通系统的自适应指数模型预测控制策略。介绍了PHEV混合动力系统及工作模式,建立了动力系统关键部件模型;提出了带SOC参考轨迹的自适应指数模型预测控制方法,在模型预测方面,使用支持向量机识别车辆行驶工况,根据行驶工况自适应调整指数模型,实现了车辆状态高精度预测;在SOC轨迹规划方面,使用智能交通系统实时获取交通信息,基于DP算法规划预测时域内SOC轨迹。经实验验证,相比于理论最优控制,基于实时交通信息的自适应指数模型预测控制方法实现了96.22%的使用经济性,无交通信息参考的模型预测控制只能实现84.34%的使用经济性,说明这里提出的控制方法可以实现很高的使用经济性。In order to lessen use cost of Plug-in Hybrid Electric Vehicle,adaptive exponent model prediction control strategy based on intelligent transportation system is proposed.Hybrid power system and its working model of PHEV are introduced,and critical component model of power system is built.Adaptive exponent model prediction control with SOC reference strategy is put forward.In aspect of prediction model,vehicle driving cycle is recognized by SVM,and exponent model can adjust adaptively depending on driving cycle,so that vehicle state prediction accuracy is improved.In aspect of SOC trajectory planning,traffic information is acquired by intelligent transportation system,and SOC trajectory in predictive step length is planned by DP algorithm.Clarified by trial,compared with optimal control in theory,adaptive exponent model predictive control based on real traffic information can realize use economy of 96.22%,and 84.34%without traffic information.It means the control method proposed by the essay can realize high use economy.

关 键 词：插电式混合动力汽车 使用经济性 自适应指数模型 智能交通系统 

分 类 号：TH16[机械工程—机械制造及自动化] TP273.1[自动化与计算机技术—检测技术与自动化装置]