State-of-Health Estimation of Lithium Battery Based on PKO-Bagging-Adaboost Ensemble Learning Algorithm

In this paper, five health factors (HFs) were extracted from the data of charge and discharge of lithium batteries. A Bagging-AdaBoost ensemble learning model was employed, which was further optimized using the PKO algorithm. The predictive accuracy of the proposed model was compared with that of PKO-SVM and PSO-Bagging-AdaBoost. Experimental results demonstrate that the PKO-Bagging-AdaBoost model achieve mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of 0.0029072,0.0036156, and 0.36563 %, respectively, thereby validating the superior performance of the proposed model.