基于PCA-GPR的锂离子电池剩余使用寿命预测

从充电过程中的电压-容量曲线中提取出一个与电池寿命高度相关健康因子（HI）。然后利用主成分分析（PCA）对影响电池寿命的多维因素进行分析和降维,结合高斯过程回归（GPR）机器学习方法提出一个基于PCA-GPR的锂离子电池剩余使用寿命预测模型。最后进行锂离子电池剩余使用寿命预测并与PCA-BP神经网络、PCA-支持向量机（SVM）模型进行比较。结果表明,利用该文提出的HI及预测模型可有效提高锂离子电池剩余使用寿命预测精度,其中通过贝叶斯优化器优化后的PCA-GPR模型的预测效果最佳。

PREDICTION OF REMAINING USEFUL LIFE OF LITHIUM-ION BATTERIES BASED ON PCA-GPR

A health indicator （HI） is extracted from the voltage-capacity curve during the charging process, which is highly correlated with the battery life. Then a PCA-GPR based prediction model for the remaining useful life of lithium-ion batteries is proposed by using principal component analysis （PCA） to analyze and downscale the multidimensional factors affecting the battery life, combined with gaussian processes regression （GPR） machine learning method. Finally, the remaining lifetime prediction of lithium-ion battery is performed and compared with PCA-BP neural network and PCA-Support vector machine （SVM） models. The results show that the HI and prediction model proposed in this paper can effectively improve the prediction accuracy of the remaining useful life of lithium-ion batteries. Among the compared prediction models, the PCA-GPR model optimized by Bayesian optimizer exhibits the best prediction performance.