混合动力电动汽车中利用决策树CART算法的能源管理方案

针对混合动力电动汽车(HEV)氮氧化物( )排放的问题,提出了一种基于决策树CART算法的柴油混合动力能源管理策略。首先,提出了一种结合决策树与回归树的分类算法(Classification and Regression Tress,CART),针对类别和变量特征,从一个或多个预测变量中预测出个例的趋势变化关系；然后,通过控制发动机和电动机之间的扭矩分配,引入了额外的自由度以调整从纯燃料经济性情况到纯 限制情况的优化权衡；最后,采用基于软件在环路和硬件在环仿真的方法,从而根据动力系统配置了解系统性能,并调整所提出的能源管理策略。实验结果表明,提出的柴油混合动力能源管理策略中, 的减少对燃料消耗的影响,且可以通过选择最佳工作点和限制发动机动力来限制 排放的潜力。相比其他几种较新的同类方案,提出的方案在同等燃料消耗的情况下 排放量更小,在燃料消耗略有下降的情况下,可以显着降低 。

A Power Management Strategy Using Decision Tree CART Algorithm for Diesel Hybrid Electric Vehicle

To solve the problem of nitrogen oxide (nox) emission of hybrid electric vehicle (HEV), a diesel hybrid energy management strategy based on decision tree CART algorithm is proposed. Firstly, a classification algorithm (Classification and Regression Tress,CART combining regression tree and decision tree is proposed. According to the characteristics of categories and variables, the trend relationship of each case is predicted from one or more predictive variables. Then, by controlling the torque distribution between the engine and the motor, additional degrees of freedom are introduced to adjust the optimization tradeoff from pure fuel economy to pure restriction; Finally, the simulation method based on software in the loop and hardware in the loop is adopted to understand the system performance according to the power system configuration and adjust the proposed energy management strategy. The experimental results show that the proposed diesel hybrid energy management strategy can reduce the impact on fuel consumption and limit the emission potential by selecting the best operating point and limiting engine power. Compared with other relatively new schemes, the proposed scheme has smaller emissions under the same fuel consumption, and can be significantly reduced under the condition of slightly reduced fuel consumption.