基于GSO-BP神经网络的磷酸铁锂电池SOC预测  

SOC prediction of LiFePO_4 cell based on GSO-BP neural network

作　　者：王丙元 张丹丹 WA NG Bingyuan;ZHA NG Dandan(College of Electronic Information and Automation,CA UC,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《中国民航大学学报》2018年第5期60-64,共5页Journal of Civil Aviation University of China

基　　金：国家自然科学基金项目(U1333102)

摘　　要：在分析磷酸铁锂电池充放电机理的基础上,建立基于BP神经网络的磷酸铁锂荷电状态(SOC)预测模型。萤火虫算法(GSO)具有不需要目标函数梯度信息、不易陷入局部最优的特点,用于优化BP神经网络的权值和阈值,可提高磷酸铁锂电池SOC预测的精度。实验结果表明:基于GSO-BP神经网络的锂电池SOC预测结果较其他神经网络预测结果更为准确,其预测结果和实际值的预测误差小于1%,符合SOC预测误差5%的技术指标要求。Based on the charge and discharge mechanisnl 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 has low probability of falling into the local opfinmm, 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 cmTent model is less than 1%, which can meet the technical requirement of SOC prediction error of 5%.

关 键 词：萤火虫优化算法 BP神经网络 磷酸铁锂电池 SOC预测 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] U469.72[自动化与计算机技术—计算机科学与技术]