Adaptive Energy Management Strategy Based on Intelligent Prediction of Driving Cycle for Plug−In Hybrid Electric Vehicle

Abstract

Under the dual−carbon goal, the research on energy conservation and emission reduction of new energy vehicles has once again become a current hotspot, and plug−in hybrid electric vehicles (PHEVs) are the first to bear the brunt. In order to improve the fuel economy of PHEV, an adaptive energy management strategy is designed on the basis of the intelligent prediction of driving cycles. Firstly, according to the vehicle dynamics model, the optimal control objective function of PHEV is established, and the relationship between vehicle fuel consumption and driving cycle is analyzed. Secondly, the initial weights and threshold of the backpropagation (BP) neural network are optimized using the particle swarm optimization (PSO) algorithm, and a PSO−BP neural network vehicle velocity prediction controller is established. Thirdly, combined with the approximate equivalent consumption minimization strategy (ECMS) algorithm to calculate the optimal initial equivalent factor in the prediction time domain, the fast−planning SOC and PI control are introduced to determine the optimal equivalent factor sequence, and the optimal torque distribution ratio of the engine and motor is calculated. Lastly, three different energy management strategies are simulated and verified under six China light−duty vehicle test cycle−passenger car (6*CLTC−P) driving cycles. Simulation results show that the established velocity prediction model has good prediction accuracy, and the proposed adaptive energy management strategy based on prediction is 9.85% higher than the rule−based strategy in terms of fuel saving rate and 5.30% higher than the ECMS strategy without prediction, which further improves the fuel saving potential of PHEV.