Global optimization plays an important role in the energy management strategies (EMS) of the hybrid electric vehicles (HEV). The fuel consumption of HEV could be reduced significantly with an acceleration of global optimization and application of global result in real-time control. In this paper, a new algorithm called breadth first search (BFS) was first used to realize the global optimization in a parallel mild HEV, which transforms the energy management problem of HEV into optimal path searching. Through simulation and calculation, it was found that the totally identical control strategies and fuel consumption could be obtained with BFS and dynamic programming (DP) respectively, while the calculation time for BFS was just about 50%-60% of that. With BFS results as reference, particle swarm optimization was used to adjust the equivalent factor in real-time and an adaptive equivalent consumption minimization strategy (A-ECMS) based on BFS was proposed. The fuel consumption could be decreased with the proposed A-ECMS by 8–15% in different driving cycles compared with that using rule-based strategies. It is believed that BFS has great potential in the future research on EMS of the HEVs.

全局优化在混合动力电动汽车 (HEV) 的能源管理策略 (EMS) 中发挥着重要作用。加速全局优化和全局结果在实时控制中的应用，可以显着降低HEV的油耗。本文首次采用广度优先搜索（BFS）新算法实现并行轻度混合动力汽车的全局优化，将混合动力汽车的能量管理问题转化为最优路径搜索。通过仿真计算发现，BFS和动态规划（DP）分别可以得到完全相同的控制策略和油耗，而BFS的计算时间仅为其50%-60%左右。以 BFS 结果为参考，采用粒子群算法实时调整等效因子，提出一种基于BFS的自适应等效消耗最小化策略(A-ECMS)。与使用基于规则的策略相比，在不同的驾驶循环中，所提出的 A-ECMS 可以将燃料消耗降低 8-15%。相信BFS在未来HEV的EMS研究中具有巨大的潜力。