Adaptive Fuzzy Logic Energy Management Strategy Based on Reasonable SOC Reference Curve for Online Control of Plug-in Hybrid Electric City Bus

The key challenge in energy management of plug-in hybrid electric vehicles is how to macroscopically plan the battery energy considering trip information while microscopically distributing the torque between power sources to improve the energy efficiency of the hybrid propulsion system. Nowadays, future partial trip information can be obtained from an intelligent transportation system and a navigation system when traveling starts. But how to use these trip information sufficiently remains to be further studied. Moreover, the online energy management strategy should not take up too many resources of micro-controller. Based on these conditions, we proposed a novel energy management strategy for plug-in hybrid electric city buses. First, Pontryagin’s minimum principle was used to obtain the optimal results of different driving cycles. Then, a neural network module was designed and trained to learn the mechanisms of optimal state-of-charge (SOC) curves. When trip starts, the neural network module can be used to generate a reasonable battery SOC reference curve according to the partial trip information available. Finally, an adaptive fuzzy logic controller was applied to follow the trend of the SOC reference curve. The main contribution of this paper is that the effectiveness of battery energy planning by comprehensively considering the future partial trip information and historical optimal SOC curves is verified, which provides a new perspective for real vehicle energy management. Simulation results show 4.61%–13.49% fuel savings on the trained and untrained driving cycles as compared with the charge-depleting and charge-sustaining strategy.