基于GRNN模型改进型对电池容量估计的研究  

Research on Battery Capacity Estimation Based on GRNN Model

作　　者：张树川[1] 孙巍 ZHANG Shuchuan;SUN Wei(School of Safety Science and Engineering,Anhui University of Science and Technology,Huainan 232001)

机构地区：[1]安徽理工大学安全科学与工程学院,淮南232001

出　　处：《长春理工大学学报（自然科学版）》2024年第6期74-82,共9页Journal of Changchun University of Science and Technology(Natural Science Edition)

基　　金：安徽省省级研究生线上线下混合示范课程(2022hhsfkc021)。

摘　　要：为了在高效率少量数据统计的情况下更为精准地估算18650电池剩余容量,基于GRNN非线性回归理论径向基神经网络模型,使用网格搜索对模型的核函数参数进行改进,使得其在少数据的情况下依旧能精准估算18650电池剩余容量。对电池进行循环充放电实验,提取循环中的相关参数,使用一节电池的数据进行训练,随后对其中4节进行容量估算,得出结果。研究表明:电池充放电时的参数,电池欧姆内阻与电池容量呈负相关,等压降放电时间与容量呈正相关。得到GRNN模型改进型,核函数参数为5,新模型对于少量数据的情况下的计算更为准确。In order to estimate the remaining capacity of 18650 batteries more accurately under the condition of high efficiency and small data statistics,based on the radial basis neural network model of GRNN nonlinear regression theory,the kernel function parameters of the model were improved by using grid search,so that the remaining capacity of 18650 batteries could still be accurately estimated under the condition of less data.The battery was charged and discharged in cycles,the relevant parameters in the cycle were extracted,and the data of Yijie battery were used for training,and then the capacity of 4 of them was estimated to obtain the results.The results show that the ohmic internal resistance of the battery is negatively correlated with the capacity of the battery,and the discharge time of iso voltage drop is positively correlated with the capacity.The improved GRNN model is obtained,and the kernel function parameter is 5,and the new model is more accurate in the case of a small amount of data.

关 键 词：少量数据情况 电池容量估算 GRNN模型 核函数参数 网格搜索 

分 类 号：X934[环境科学与工程—安全科学]