基于工况预测的PHEV自适应等效油耗最小策略

Adaptive Equivalent Fuel Consumption Minimization Strategy for PHEV Based on the Prediction of Vehicle Driving Conditions

为解决等效燃油消耗最小策略(ECMS)等效因子自适应更新的问题,采用前向建模的方法建立了插电式并联混动汽车整车模型。首先基于EMD-NAR神经网络进行未来车速的预测,并基于K-means聚类分析的方法建立了工况识别模型,通过判断当前预测工况所属类别,实现等效因子的自适应调节。在标准工况下,对基于工况预测的自适应等效燃油最小策略与全局最优控制策略DP以及等效燃油消耗最小策略的各项性能参数进行了仿真对比,验证了该控制策略的有效性和等效因子选取方法的可行性。

In order to solve the adaptive updating of equivalent factors of the equivalent consumption minimization strategy(ECMS),a full vehicle model of plug-in parallel hybrid electric vehicle was established with the method of the forward modeling.The future speed was predicted based on EMD-NAR neural network,and the vehicle driving cycle recognition model was established based on K-means clustering analysis.The adaptive adjustment of equivalent factors was realized by judging the category of the current predicted working condition.The performance parameters of the adaptive equivalent consumption minimization strategy based on condition predictions,the global optimal control strategy DP and the equivalent consumption minimization strategy are simulated and compared in the standard driving conditions,which verifies the effectiveness of the control strategy and the feasibility of the method for the selection of the equivalent factors.