农机使用与维修Issue(9)：1-8,8.DOI:10.14031/j.cnki.njwx.2025.09.001
基于LSTM下新能源拖拉机电池健康状态预测研究
Research on Battery Health State Prediction of New Energy Tractor Based on LSTM
摘要
Abstract
The health state of power battery of new energy tractor directly affects the operation efficiency and service life,and accurate prediction of battery SOH is of great significance to improve the performance of battery management system,optimize the battery maintenance strategy,and guarantee the stability of agricultural production.In this paper,based on the grey correlation analysis of the correlation between the extracted features and the degradation of the battery capacity and the sorting,we screened out eight highly correlated features,such as the time of the maximum charging temperature and the maximum discharging temperature.eight features with high correlation,such as the maximum value of discharge temperature,and then construct a long and short-term memory network prediction model,optimise the data quality by using data preprocessing techniques,and carry out SOH prediction based on the LSTM deep learning model.The experi-mental results show that the method has higher prediction accuracy and generalisation ability compared with the tradition-al support vector regression and random forest methods.This study provides an efficient and intelligent means of predic-ting the battery health management of new energy agricultural equipment,aiming to further enhance the intelligent level of agricultural production and energy use efficiency.关键词
新能源拖拉机/电池健康状态/长短时记忆网络/预测模型/农业智能化Key words
new energy tractor/State of Health(SOH)/Long Short-Term Memory(LSTM)/prediction model/agri-cultural intelligence分类
农业科技引用本文复制引用
沈晨普,李辉,刘进福..基于LSTM下新能源拖拉机电池健康状态预测研究[J].农机使用与维修,2025,(9):1-8,8.基金项目
江苏省自然科学基金(BK20220241) （BK20220241）
江苏省高职院校教师访学研修项目(2024TDFX004) （2024TDFX004）
2024年江苏省青蓝工程优秀青年骨干教师项目 （）
第四批常州市领军型创新人才项目(CQ2021079) （CQ2021079）
常州大学高等职业教育研究项目(CDGZ202530) （CDGZ202530）
常州市基础研究计划(应用基础研究)面上项目(CJ20220181) （应用基础研究）
