A DOD-SOH balancing control method for dynamic reconfigurable battery systems based on DQN algorithm

Abstract

This article presents a DOD-SOH equalization method for a DRB system based on the Deep DQN algorithm. The proposed method utilizes DQN to learn the operational processes of the system. By integrating the advantages of DRB with SOH equalization theory and the DQN algorithm from the perspective of DOD, our method significantly improve battery performance and ensure cell balancing. To begin with, we present a dynamic reconfigurable battery system with a simple topological structure and outline its switching control process. Additionly, we provide an analysis of the SOH balancing principle and elaborate on the control process of DQN algorithm. Finally, subsequent simulations are carried out, and the simulation results demonstrate outstanding performances in reducing the variance of SOHs, which indicates an enhancement in the level of SOH balancing as well.