基于混杂模型预测控制的混合动力车辆能量管理  

Energy management of HEV based on hybrid model predictive control

作　　者：江冬冬[1] 李道飞[1] 俞小莉[1] JIANG Dong-dong;LI Daofei;YU Xiaoli(Institute of Power Machinery and Vehicular Engineering,Zhejiang University,Hangzhou 310027,China)

机构地区：[1]浙江大学动力机械及车辆工程研究所,杭州310027

出　　处：《吉林大学学报（工学版）》2020年第4期1217-1226,共10页Journal of Jilin University:Engineering and Technology Edition

基　　金：浙江科技厅重点研发计划项目(2018C01057,2018C01058);中央高校基本科研业务费专项项目(2019QNA4018);宁波市“科技创新2025”重大专项(2018B10063,2018B10064)。

摘　　要：以并联式混合动力车辆为研究对象,针对其具有连续变量和离散变量并存的特点,采用混杂模型预测控制算法对其进行优化求解。首先,对发动机和电机的外特性曲线、行驶阻力曲线、发动机油耗曲面、电池功率曲面等进行线性化处理,并采用分段仿射自回归外生模型来拟合电池荷电状态计算公式;其次,在混杂系统描述语言工具箱中对各个线性化部分进行描述,并自动生成混合逻辑动态系统;最后,采用混杂工具箱对生成的混合逻辑动态系统进行求解,并将结果与基于非线性模型预测控制算法的结果进行对比。对比结果表明:采用这两种算法对UDDS、NEDC和US06这3个标准工况进行仿真时,基于混杂模型预测控制算法相对于非线性模型预测控制算法,燃油经济性分别提高了20.0%、15.5%和10.6%,且当预测时域小于9时,混杂模型预测控制算法的计算速度更快。Considering that the parallel hybrid electric vehicle contains both continuous variables and discrete variables, a hybrid model predictive control algorithm is adopted to optimize the energy management. Firstly,the external characteristic curve of engine and motor,driving resistance curve,engine fuel consumption surface,battery power surface are processed by piece-wise affine linearization.Then,the battery state of charge formula is fitted by piecewise auto-regressive exogenous models. Finally,each linearized part is described in the hybrid system description language toolbox,which can automatically generate and solve the mixed logical dynamical system. The simulation results of hybrid model predictive control under the UDDS,NEDC and US06 standard operating conditions are compared with the nonlinear model predictive control algorithm. It is shown that the fuel economy of hybrid model predictive control algorithm is 20.0%、15.5% and 10.6% higher than that of the nonlinear model predictive control algorithm,respectively. If the prediction horizon is less than 9,the computation speed of hybrid model predictive control algorithm is faster.

关 键 词：车辆工程 并联式混合动力车辆 能量管理 混杂模型预测控制 混合逻辑动态系统 非线性模型预测控制 

分 类 号：U469.72[机械工程—车辆工程]