基于神经网络和DP算法的PHEV能量管理策略研究

Research on PHEV Energy Management Strategy Based on Neural Network and DP Algorithm

以某款PHEV为研究对象,利用DP算法求得UDDS工况下离线最优控制序列。利用神经网络对离线最优控制序列及对应汽车状态进行学习训练,建立控制变量神经网络在线预测模型。仿真结果表明,所设计策略实现了离线最优控制变量的良好跟随,且燃油经济性提升明显。

Taking a PHEV as the research object,the DP algorithm is used to obtain the offl ine optimal control sequence under UDDS conditions.The article uses the neural network to learn and train the offl ine optimal control sequence and the corresponding car state,and establish an online prediction model of the control variable neural network.The simulation results show that the designed strategy achieves a good follow-up of the offl ine optimal control variables,and fuel economy is signifi cantly improved.