基于GWO-SVR的锂电池剩余使用寿命预测

Prediction of Remaining Useful Life of Lithium-ion Batteries Based on GWO-SVR Method

锂离子电池剩余使用寿命预测在电池管理系统中发挥着重要作用,准确预测其剩余使用寿命能够保障电池的安全稳定运行。由于支持向量回归SVR(support vector regression)参数内核选择较为困难,为此提出灰狼优化—支持向量回归GWO-SVR(gray wolf optimization-SVR)方法,使用灰狼算法优化其内核参数,根据NASA预测中心提供的电池数据集对该方法进行了验证。通过与SVR方法进行对比发现,所提GWO-SVR方法的预测精度得到显著提高;在此基础上与ALO-SVR方法进行对比,证明所提方法平均相对误差降低了7.16%,预测精度更高,有效地提高了锂离子电池剩余寿命预测的精确性。

The prediction of remaining useful life(RUL)of lithium-ion batteries plays an important role in the battery management system,and an accurate prediction can ensure the safe and stable operation of batteries.Due to the difficulty in selecting the parameter kernel of support vector regression(SVR),a gray wolf optimization(GWO)method for SVR is proposed,and the gray wolf algorithm is used to optimize the kernel parameters.This method is verified according to the battery dataset provided by the NASA Prognostics Center of Excellence.Compared with that of the SVR method,the prediction accuracy of the proposed GWO-SVR method is significantly improved.On this basis,the proposed method is further compared with the Ant Lion Optimizer(ALO)-SVR method,and it is found that the average relative error is reduced by 7.16%while the prediction accuracy is higher,indicating that the prediction accuracy of RUL is effectively improved.