In order to improve the accuracy of the GRU neural network model in predicting the remaining useful life(RUL)of lithium-ion batteries,the GRU model was optimized based on PCA-GWO and then applied in the prediction.The results show that compared with the traditional GRU model,the PCA-GWO-GRU model presents higher prediction accuracy.When the starting point of the prediction is 90%of the original data,the prediction accuracy can reach the highest,with the corresponding RMSE of 0.004 9,MAE of 0.003 6,and R2 of 0.986 3.
关键词
锂离子电池/剩余使用寿命预测/GRU/灰狼算法/主成分分析
Key words
lithium-ion battery/remaining useful life(RUL)prediction/GRU/gray wolf optimizer(GWO)/principal component analysis(PCA)