An Intelligent Energy Management Strategy for Plug-in Hybrid Electric Vehicle Inspired from Monte Carlo Tree Search

Abstract

The equivalent consumption minimisation strategy (ECMS) solves the problem that dynamic programming (DP) cannot be applied in real-time. Its equivalence factor balances the consumption of different power sources and affects the maintenance of battery state of charge (SOC) of plug-in hybrid electric vehicles in driving cycles. However, a systematic optimisation with known driving cycles gives the optimal choice of the equivalence factor. This paper proposed an intelligent energy management strategy (IEMS), which dynamically adjusts the equivalence factor according to the power demand and SOC. IEMS introduces a hierarchical control architecture: the equivalence factors are given by the planning results of the Monte Carlo tree search (MCTS) in the upper layer; ECMS calculates the energy distribution in the bottom layer. A backward parallel plug-in hybrid electric vehicle model was used to train the IEMS. The experiment showed that under the complete driving cycles, the fuel consumption of IEMS can be close to DP. In terms of fuel economy, it exceeds ECMS with a constant equivalence factor and Q-learning.