Energy Management Strategy for Hybrid Electric Vehicle Based on Driving Condition Identification Using KGA-Means

Abstract

In order to solve the problem related to adaptive energy management strategies based on driving condition identification being difficult to be applied to a real hybrid electric vehicle (HEV) controller, this paper proposes an energy management strategy by combining the driving condition identification algorithm based on genetic optimized K-means clustering algorithm (KGA-means), and the equivalent consumption minimization strategy (ECMS). The simulation results show that compared with ECMS, the energy management strategy proposed in this article drives the engine working point closer to the best efficiency curve, and smooths out the state of charge (SOC) change and better maintains the SOC in a highly efficient area. As a result, the vehicle fuel consumption reduces by 6.84%.