基于行驶工况识别的混合动力货车能量管理策略

Energy management strategy of hybrid electric truck based on driving condition recognition

针对车辆行驶工况对混合动力汽车的燃油经济性有较大影响的问题,以一款行星混联式混合动力载货汽车为研究对象,结合工况识别方法、动态规划算法以及神经网络构建整车的能量管理策略.利用动态规划算法在4种典型工况下的最优功率分配数据训练神经网络,分别得到适应于各典型工况的神经网络模型.在控制过程中利用工况识别方法对行驶工况进行识别,根据工况当前所属的类别,选择相应的神经网络模型对需求功率在动力源间的分配进行实时优化.仿真结果表明:与工况识别前的能量管理策略相比,采用所提出的能量管理策略时车辆燃油消耗降低了6.25%,同时电池SOC更加稳定,有利于电池的使用寿命.

Aiming at the problem that the driving conditions have a great influence on the fuel economy of the hybrid electric vehicle,a planetary hybrid truck is taken as the research object,the energy management strategy of the vehicle is established by combining driving condition recognition method,dynamic programming algorithm and neural network.The neural network is trained by using the optimal power distribution data of dynamic programming algorithm under four typical driving conditions,and the neural network models which are respectively suitable for each typical driving condition are obtained.During the control process,driving conditions are identified by the condition identification method.According to the current category of driving conditions,the corresponding neural network model is selected to optimize the distribution of demand power among power sources in real time.The simulation results show that compared with the energy management strategy before condition recognition,the vehicle fuel consumption decreases by 6.25%,and the SOC of the battery is more stable,which is beneficial to the battery lifespan.