基于DIRECT算法的混合动力汽车参数优化研究

Study on Parameters Optimization of Hybrid Electric Vehicle Based on DIRECT Algorithm

研究混合动力汽车参数优化的方法。建立了以油耗最小为目标,以动力性设计指标为约束,以发动机功率、电机功率、电池容量、电池总线电压为变量的混合动力汽车部件参数优化模型,并应用分割矩形(DIRECT)算法,对此模型进行求解,找到了油耗最小的部件参数,进行了仿真验证。结果表明,在EPA城市循环工况(UDDS)和高速公路循环工况(HWFET)下,经过该方法优化后的混合动力汽车百公里油耗分别降低9.3%和14%。

An optimization of the power train parameters of a hybrid electric vehicle has been studied on. The optimization model aiming at minimizing the fuel consumption with the constraint of dynamic perlbrmance design was built. The DIRECT （Dividing Rectangle） algorithm has been used to find the global optimal parameters including engine power, motor power, battery capacity and battery bus voltage. Tile results show that the fuel consumption after optimization is 9.3% and 14% respectively less than that before optimization under two typical driving cycles of the EPA Urban Dynamometer Driving Schedule （UDDS） and Highway Fuel Economy Test （HWFET）.