基于BP算法的氢镍电池充电均衡策略

Equalization strategy for MH-Ni battery charging based on BP algorithm

为了提高电动车的使用寿命,改善电池组的工作性能,在对典型的电压均衡控制和荷电状态均衡控制进行理论分析的基础上,提出了基于变步长的反向传播误差(back propagation,BP)神经网络及200 Ah氢镍电池小电流均衡控制策略。该均衡方法采用大电流充电和小电流均衡,考虑了充电过程中的析气与复合。对电池组进行的模拟仿真结果表明,电压检测精度得到提高,具有良好的均衡效果和均衡速度。

In order to extend the electrical vehicle （EV） life span and to improve battery pack working performance, BP neural network based on variable step size and equalization control method for trickle charge 200Ah MH-Ni battery was proposed. Then theoretical analysis of typical voltage equalization control and SOC equalization control was particularly described. Bulk charge and small current equalization were both applied to this equalization system. Gassing and recombination were both taken into consideration. Simulations of battery pack were conducted. Conclusions show that voltage detection accuracy is improved. Both equalization efficiency and speed were fine.