A Real-Time Energy Management Strategy Based on Energy Prediction for Parallel Hybrid Electric Vehicles

Hybrid electric vehicle (HEV) technology is an effective way to resolve the problems of energy consumption and air pollution. Energy management strategies are critical to the performance of HEVs. In this paper, a novel energy management strategy of equivalent consumption minimization strategy (ECMS)-type is proposed for parallel HEVs based on energy prediction (ECMS-EP). The energy prediction is estimated based on the predicted velocity that is calculated by a chaining-neural-network method over different temporal horizons. A novel adaptive rule has been developed by eliminating the need to reset the initial equivalent factor (EF) based on the energy prediction to adjust the EF of ECMS-EP in real time. The control objective is to improve the fuel economy and sustain the state of charge (SoC). Then, via MATLAB/Simulink, simulations are conducted in three different prediction horizon lengths to verify the performance of the proposed ECMS-EP with adaptive rules. The simulation results show that the proposed ECMS-EP is able to achieve more stable SoC trajectories and better fuel economy with a fuel consumption reduction of 2.7%-7% compared with the traditional adaptive-ECMS.