混合动力汽车多目标参数解耦优化方法研究

混合动力汽车的能量管理策略与动力总成参数高度耦合,在混合动力汽车优化过程中存在不断循环的情况,使得参数优化难以实现,无法找到最优解。针对这一问题,提出一种多参数解耦优化方法,该方法采用混合优化策略,将动力性目标作为约束条件,以粒子群算法优化动力总成参数,并通过粒子群算法对不同参数下的能量管理策略与换挡策略进行瞬时优化。针对一并联混合动力汽车,利用Matlab/Simulink建立了包含模糊PID驾驶员在内的用于优化过程中自适应的正向仿真模型。结果表明,混合优化算法能最大程度挖掘该动力总成的潜力,相较于同时优化逻辑门阈值的优化方法,经济性目标提高了4. 55%,并能同时得到对应该最优参数的具有通用性的能量管理策略和换挡策略。

Research of Multi-objective Parameter Decoupled Optimization Method for Hybrid Electric Vehicle

The energy management strategy (EMS) and the powertrain parameter of hybrid electric vehi cles are highly coupled. In the process of hybrid electric vehicle optimization, there is a continuous cycle, which makes it difficult to achieve parameter optimization and find the optimal solution. Aiming at this prob lem, a multi-parameter decouped optimization method is proposed, which adopts hybrid optimization strategy, taking the dynamic targets as constraint conditions and using particle swarm optimization algorithm to opti mize powertrain parameter, the Particle swarm optimization (PSO) is used to optimize the energy management strategy and shifting strategy under different parameters. Aiming at a parallel hybrid vehicle, a forward model that includes a fuzzy PID driver is established by using Matlab/Simulink to self-adapt to the changing power train configurations. The results show that the hybrid optimization methodology is able to squeeze the poten tial of the vehicle, compared with the optimization method that simultaneously optimizes the logic gate thresh old, the economy performance is enhanced by 4.55% and simultaneously, the energy management strategy and shifting strategy for the HEV under these parameters are obtained.