摘要
利用Elman神经网络完成了对锂电池SOH的估算,以温度、SOC和内阻作为Elman神经网络模型的输入,SOH值作为输出建立Elman神经网络模型,并对模型进行测试,仿真结果表明:该模型的SOH测试相对误差在±1.5%内.最后采用MCU实现数据的采集,在服务器中用java实现基于Elman神经网络的估算算法,该方法释放了下位机的运算压力,提高了处理速度和精度.
The Elman neural network is used to estimate the SOH of a lithium battery.The temperature,SOC and internal resistance are used as the input of the Elman neural network model,and the SOH value is used as the output to establish the neural network model.The accuracy of the model is tested,and the relative error is within±1.5%,which achieves a high test accuracy.The MCU is used in data acquisition,and the algorithm based on Elman neural network is implemented in the server.This method releases the calculation pressure of the lower computer and improves the processing speed and accuracy.
作者
邹娟
徐升荣
曾洁
ZOU Juan;XU Shengrong;ZENG Jie(School of Materials Science and Engineering,Dalian Jiaotong University,Dalian 116028,China;Jiangsu Provincial Electric Power Co.,Ltd,Suqian Branch,Suqian 223800,China)
出处
《大连交通大学学报》
CAS
2020年第2期104-108,共5页
Journal of Dalian Jiaotong University
基金
辽宁省自然科学基金资助项目(20180551121)
辽宁省教育厅科学研究计划资助项目(JDL2017039)。
关键词
锂电池
SOH估算
ELMAN神经网络
特征参数
Lithium battery
SOH estimation
elman neural network
characteristic parameters
