基于动态规划的自适应等效最小燃油消耗策略

Adaptive Equivalent Consumption Minimization Strategy Based on Dynamic Programming

针对等效燃油消耗最小策略（ECMS）中最优等效因子对工况依赖性较大的问题，以单轴并联式混合动力汽车为研究对象，以动态规划算法（DP）获得的特定行驶工况下最优控制结果为基础，逆推得到ECMS的最优等效因子；拟合出平均最优等效因子随电量维持水平的变化规律，以此为基础设计了一种自适应等效燃油消耗最小策略．在中国典型城市工况下的仿真结果表明：对于任意初始SOC值，文中策略能够较好地维持电量平衡；在SOC稳定之后的油耗为828g，与DP对比仅有0．8％的误差．多种不同工况仿真结果表明，SOC变化规律与中国典型城市工况仿真结果一致，在SOC稳定之后，单个循环工况的油耗相对DP仅增加了0．1％～0．2％．

Aiming at the problem that the optimum equivalent factor relies heavity on the driving cycle in the equiva- lent consumption minimization strategy （ECMS） , by taking a single-axis parallel hybrid electric vehicle as the re- search object and based on the optimal control results obtained through the dynamic programming （DP） in a special driving cycle, an instantaneous optimal equivalent factor in the framework of the ECMS is obtained in a reverse way. Then, the change rules of the average optimai equivalent factor with the charge-sustaining level are fitted. On this basis, an adaptive equivalent consumption minimization strategy is designed. The simulation results of a typical urban driving cycle in China show that, for an arbitrary initial SOC, the designed strategy can stabilize the battery capacity in a fixed value, and that the fuel consumption is 828 g after stabilizing the SOC, which indicates an error of only 0. 8 % in comparison with that obtained through the DP. The simulation results of other kinds of driving cy- cles show that the change rules of the SOC in the other cycles accord well with those in the typical urban driving cy- cle, and after stabilizing the SOC, the fuel consumption increases only by 0. 1% - 0. 2% in comparison with that obtained through the DP in a single cycle.