Rule-Based Energy Management Strategy for a Power-Split Hybrid Electric Vehicle with LSTM Network Prediction Model

Nowadays, automotive manufacturers have led to the rapid development of hybrid electric vehicles to improve fuel economy and emission performance. In hybrid electric vehicles, the energy management strategy is crucial since it determines the power flow pattern and significantly affects vehicle performance. Therefore, in this paper, two rule-based control strategies, i.e., Engine-Dominant strategy and Motor-Dominant strategy, are proposed for a power-split configuration and compared in terms of fuel consumption and emissions under a city-highway combined driving cycle. Then, a long short-term memory recurrent neural network is designed to predict the control variables. Based on simulation results, the proposed model can provide reasonable predictions with acceptable deviations. Moreover, compared to the baseline controller, a 14.5% improvement in fuel economy is observed with the predicted data in a highway driving cycle.