关键词: 萤火虫优化算法, BP 神经网络, 磷酸铁锂电池, SOC 预测

Abstract: Based on the charge and discharge mechanism of LiFePO4 cell, a prediction model is developed for the SOC (state of charge) of Lithium iron phosphate battery based on BP neural network. As GSO (glowworm swarm optimization) does not require gradient information of the objective function，and it is not easy to fall into the local optimum, it can be used to optimize the weight and threshold of BP neural network and to improve SOC prediction accuracy of lithium iron phosphate battery. Experimental results show that SOC prediction of lithium battery based on GSO-BP neural network is more accurate than that of other neural networks, prediction error of the current model is less than 1%, which can meet the technical requirement of SOC prediction error of 5%.

Key words: GSO, BP neural network, LiFePO4 cell, SOC prediction