新能源汽车主要以锂离子电池作为储能单元,随着锂离子电池的老化,在充电过程中极易发生燃烧甚至爆炸,给电网和充电场站带来巨大安全隐患.电池健康状态(State of Health,SOH)是评价电池老化和性能衰减的重要指标.通过研究锂离子电池的充电特性,从中提取反映电池老化的特征向量,采用支持向量机(Support Vector Machine,SVM)估计电池的健康状态,并采用模拟退火算法优化SVM的超参数.研究表明,采用优化后的SVM预测SOH,决定系数为99.73%,预测误差小于0.05%,达到了较高的预测精度.研究成果可以为充电场站的安全运维提供锂离子电池的监测参数,提高充电过程的安全性能.
State of Health Estimation of Lithium-Ion Batteries Based on Support Vector Machine and Simulated Annealing
Lithium-ion batteries are the primary energy storage units for new energy vehicles.As lithium-ion batteries age,they become prone to combustion or even explosion during the charging process,posing significant safety hazards to the power grid and charging stations.The State of Health(SOH)of a battery is an important indicator for evaluating battery aging and performance degradation.This study examines the charging characteristics of lithium-ion batteries to extract feature vectors that reflect battery aging.A Support Vector Machine(SVM)is employed to estimate the battery's SOH,with its hyperparameters optimized using the Simulated Annealing(SA)algorithm.The research findings show that the optimized SVM achieves a determination coefficient of 99.73%and a prediction error of less than 0.05%for SOH prediction,indicating high predictive accuracy.These results can provide monitoring parameters for the safe operation and maintenance of charging stations,thereby enhancing the safety of the charging process.
NETL
NSTL
万方数据
