基于健康特征筛选与GWO-LSSVM的锂电池健康状态预测

Prediction of state of health for lithium battery based on health feature screening and GWO-LSSVM

锂电池健康状态(SOH)预测是电池管理系统(BMS)最重要的功能之一,准确有效地预测锂电池SOH可有效提升设备利用率,保证系统稳定性。为了提高预测准确度,本文提出一种基于健康特征筛选与灰狼优化算法(GWO)-最小二乘支持向量机(LSSVM)的锂电池SOH预测方法,首先采用灰色关联分析(GRA)法计算每个健康特征相对于锂电池SOH的灰色关联度,并将灰色关联度进行排序,确定SOH预测的主要健康特征;然后针对LSSVM模型参数需靠人为经验选择的问题,采用寻优性能较好的灰狼优化算法对其进行优化选择并构建GWO-LSSVM模型;最后基于NASA数据集,对模型进行训练和测试,并与其他3种模型的预测结果进行对比,对比结果证明了本文所提方法的有效性。

State of health(SOH)prediction for lithium battery is one of the most important functions of battery management system(BMS).Accurate and effective prediction of lithium battery SOH can effectively improve the utilization rate of equipment and ensure system stability.In order to improve the accuracy of prediction,this paper proposes a SOH prediction method for lithium batteries based on health feature screening and grey wolf optimizer(GWO)-least square support vector machine(LSSVM).Firstly,grey relational analysis(GRA)is used to calculate the grey relational degrees of each health feature relative to the SOH of lithium batteries,and the grey correlation degrees are sorted to determine the main health characteristics of SOH prediction.Then,aiming at the problem that the parameters of LSSVM model need to be selected by human experience,the grey wolf optimization with good optimization performance is used to optimize the parameters and build the GWO-LSSVM model.Finally,the model is trained and tested on the basis of NASA data set,and the evaluation index values of back propagation(BP),LSSVM and particle swarm optimization(PSO)-LSSVM models are compared and analyzed to prove the effectiveness of the proposed method.