Investigating adaptive-ECMS with velocity forecast ability for hybrid electric vehicles

Energy management strategy is crucial in improving the fuel economy of hybrid electric vehicles (HEVs). This paper targets at evaluating the role of velocity forecast in the adaptive equivalent consumption minimization strategies (ECMS) for HEVs. A neural network based velocity predictor is constructed to forecast the short-term future driving behaviors by learning from history data. Then the velocity predictor is combined with adaptive-ECMS to provide temporary driving information for real-time equivalence factor (EF) adaptation. Compared with traditional adaptive-ECMS, which uses historical driving profile for EF estimation, the proposed strategy is able to foresee the change of the driving behaviors and adjust the EF more reasonably. Simulation results show that, compared with traditional adaptive-ECMS, the proposed improvement with velocity forecast incorporated is able to achieve better fuel economy and more stable battery state of charge (SOC) trajectory, with a fuel consumption reduction by over 3%.