返回
Active equalization control method for battery pack based on Double-DQN
DOI:10.1016/j.est.2024.111361.png)
摘要
En 中文
In response to the issues of inconsistency and component loss during battery balancing, an active balancing scheme has been designed to control the battery balancing process. Firstly, a balancing circuit based on redundant batteries is proposed, which leverages the principle of single-cell parallel voltage division to enhance the consistency of the battery pack. To mitigate the problem of component loss, this study proposes a DoubleDQN-based balancing control strategy that effectively reduces the losses during battery balancing by considering the impact of the battery pack's state of charge (SOC) and state of health (SOH). Simulation results demonstrate that the proposed scheme not only reduces the inconsistency of the battery pack but also significantly decreases the component losses. Additionally, a hardware prototype has been developed to evaluate the efficacy of the proposed balancing scheme.
Keyword:
Machine learning
Active equalization
Redundant battery
Optimum control
期刊
IF:
9.8
论文数:
2.3W
被引数:
10.1W
机构
引用论文
Performance Evaluation of Modularized Global Equalization System for Lithium-Ion Battery Packs模块化锂离子电池组全局均衡系统性能评价
Internal short circuit detection for lithium-ion battery pack with parallel-series hybrid connections并联-串联混合连接的锂离子电池组内部短路检测
A novel dual-scale cell state-of-charge estimation approach for series-connected battery pack used in electric vehicles一种新型电动汽车串联电池组双尺度电池荷电状态估计方法
Does modified Otago Exercise Program improves balance in older people? A systematic review改良的奥塔哥锻炼计划能改善老年人的平衡吗?一项系统综述
Zero-Current Switching Switched-Capacitor Zero-Voltage-Gap Automatic Equalization System for Series Battery String串联电池组零电流开关开关电容零电压间隙自动均衡系统
Fractional-order modeling and SOC estimation of lithium-ion battery considering capacity loss考虑容量损失的锂离子电池分数阶建模与SOC估计

