[Abstract] Combining batteries and supercapacitors into a composite power supply system and combining it with effective energy management strategies can significantly improve energy utilization and extend the service life of the energy storage system. In order to minimize the energy consumption loss of the composite power system, an energy management strategy based on the dual-delay deep deterministic policy gradient (TD3) algorithm is designed. Compared with the deep deterministic policy gradient (DDPG) algorithm, this algorithm solves the problem of overestimation of the Q value and has smaller energy consumption loss. Using the electric vehicle driving equation and the equivalent circuit model of the composite power supply system, a MATLAB/Simulink simulation model based on the TD3 algorithm was built and tested. The simulation results show that the proposed energy management strategy can reduce the impact of large current on the battery. Compared with the DDPG algorithm, the energy utilization rate is increased by 1.36%, the battery peak current output is reduced by 14.68%, and the battery temperature rise is reduced by 3.52%. The total system energy consumption was reduced by 2.17%.
[Keywords]Electric vehicles; hybrid power systems; energy management; deep reinforcement learning
0 Preface
With the shortage of oil resources and abnormal climate changes, the development of pure electric vehicles has attracted more and more attention in the past few decades. Compared with fuel vehicles, pure electric vehicles have the advantages of zero pollution and low noise. As the main energy source of pure electric vehicles, batteries have the advantages of high energy density, small self-discharge, and no memory effect. However, under the condition of high-rate current charging and discharging, the battery capacity attenuation speed increases, resulting in shortened battery cycle life, and super Capacitors have the characteristics of high specific power and can be charged and discharged quickly. Therefore, combining the two into a composite power system, combined with effective energy management strategies, can give full play to the advantages of the high energy density of batteries and the high power density of supercapacitors. Energy management strategy is one of the key technologies. Currently, the main management strategies include rule-based strategies, optimization-based strategies and learning-based strategies.
Rule-based strategies include deterministic rules and fuzzy rules. Zhou Meilan and others studied logic threshold control based on deterministic rules. The main idea of this strategy is that the battery provides power when the car is driving.