湖北汽车工业学院学报2025，Vol.39Issue(2)：25-29,34,6.DOI:10.3969/j.issn.1008-5483.2025.02.005
基于Transformer编码器的电动汽车低压蓄电池SOC预测
State of Charge Prediction for Low-voltage Batteries in Electric Vehicles Based on Transformer Encoder
摘要
Abstract
To facilitate the prediction of the remaining capacity of low-voltage batteries in electric vehi-cles during operation,a method for state of charge(SOC)prediction for low-voltage batteries in electric vehicles based on the Transformer encoder was proposed.Real vehicle data was collected through a re-mote data acquisition platform.The K-nearest neighbors(KNN)algorithm was utilized to impute miss-ing values and eliminate outliers.A battery capacity feature database was subsequently established un-der scenarios of intelligent charging and static discharging.Battery capacity characteristics were ana-lyzed,and comprehensive experiments were conducted.Experimental results show that the prediction error of the proposed model is less than 2%.关键词
Transformer编码器/电动汽车/低压蓄电池/容量特征/SOC预测Key words
transformer encoder/electric vehicle/low-voltage battery/capacity characteristic/SOC prediction分类
交通运输引用本文复制引用
陈少辉,王思山,司华超,周昊夫,童乐言..基于Transformer编码器的电动汽车低压蓄电池SOC预测[J].湖北汽车工业学院学报,2025,39(2):25-29,34,6.基金项目
湖北省重点研发项目(2023BAB169) （2023BAB169）
