基于BP神经网络法估算动力电池SOC

精确估计电动汽车用动力锂离子电池荷电状态(SOC)对于电动汽车的续航里程的估计和动力电池的安全保护具有重要的意义。针对锂离子电池的非线性关系,采用BP神经网络法来估算SOC。以3.2 V/100 Ah的磷酸锂铁电池为研究对象,在恒温条件下采用Arbin BT2000系列的充放电测试仪进行充放电实验采集原始数据,并将数据导入到神经网络模型中去训练和验证。验证结果表明:用BP神经网络法估算SOC的误差能控制在5%以内,验证了模型的准确性,为相似的SOC估计算法的改进提供参考和依据。

Estimation and simulation of power battery SOC based on BP neural network

Accurately estimate the state of charge (SOC) of power lithium ion batteries for electric vehicles was of great significance to the estimation of the endurance mileage of electric vehicles and the safety protection of the power batteny.In view of the nonlinear relation of the lithium ion battery,the BP neural network method was used to estimate the SOC.The research object was lithium iron phosphate battery (3.2 V/100 Ah).The charge and discharge experiments were done by the charge discharge tester ArbinBT2000 under the constant temperature,and the raw data was collected.Finally the data was imported into the neural network model to train and verify.The validation results show that the error of SOC can be controlled within 5％ by the BP neural network method.The accuracy of the model was verified,which provided reference and basis for the improvement of similar SOC estimation algorithm.