基于RFF及GWO-PF的锂电池SOC估计

为了提高锂电池剩余电量估计的准确性,提出一种在线参数辨识与改进粒子滤波算法相结合的锂电池SOC估计方法。针对粒子滤波中的粒子退化问题,引入灰狼算法,利用灰狼算法较强的全局寻优能力优化粒子分布,保证粒子多样性,有效抑制粒子退化现象,提高滤波精度。采用带遗忘因子的递推最小二乘法实时更新模型参数,并与改进粒子滤波算法交替运行,进一步提高SOC的估计精度。实验结果表明,改进算法的平均估计误差始终保持在±0.15%以内,相比扩展卡尔曼滤波与无迹卡尔曼滤波算法,在电池SOC估计上有更高的估计精度与稳定性。

SOC Estimation of Lithium Batteries Based on RFF and GWO-PF

In order to improve the accuracy of estimation of residual charge of lithium battery, a method of estimation of SOC of lithium battery based on online parameter identification and improved particle filter algorithm is proposed. Aiming at the problem of particle degradation in particle filtering, gray wolf optimization is introduced to optimize particle distribution with its strong global optimization ability to ensure particle diversity, effectively suppress particle degradation, and improve the filtering accuracy. The recursive least square method with forgetting factor is used to update the model parameters in real-time and run alternately with the improved particle filter algorithm to further improve the estimation accuracy of SOC. The experimental results show that the average estimation error of the improved algorithm is always less than ±0.15%. Compared with the extended Kalman filter and unscented Kalman filter, the improved algorithm has higher estimation accuracy and stability in the estimation of battery SOC.