For the remaining useful life prediction of lithium-ion batteries, the reliability of the features and the validity of the regression algorithm used to construct the prediction model are very important to the prediction results. For this reason, this paper proposes a prediction method based on AdaBoost-support vector regression. First, 9 features are extracted from the battery aging data, and the correlation between features and RUL is verified. Then, the random forest is used to select the extracted features to improve the reliability of the features. Finally, based on the selected features, the prediction model of RUL is established by using AdaBoost to optimize the support vector regression model. The validity of the proposed method is verified in NASA lithium battery data set.
Keywords Lithium-ion Battery, Remaining Useful Life, Random Forest, AdaBoost, Support Vector Regression