Capacity degradation prediction of lithium-ion battery based on artificial bee colony and multi-kernel support vector regression

Abstract

Lithium-ion battery (LIB) capacity degradation prediction plays an important role in the prediction of battery health degradation. Accurate prediction of its capacity can guide battery replacement and maintenance, and ensure the safety and stability of the energy storage system. In this paper, a hybrid method based on artificial bee colony (ABC) algorithm and multi-kernel support vector regression (MK-SVR) is proposed to predict the capacity degradation of LIB. Firstly, the capacity degradation prediction model of LIB is established by MK-SVR with few hyperparameters. Then, the hyperparameters of degradation prediction model are optimized by the ABC to improve the accuracy of prediction. Finally, the proposed method is verified by NASA's LIB degradation experiment. Compared with different optimization algorithms, ABC greatly improves the accuracy of capacity degradation prediction. Compared with traditional forecasting methods, the proposed method can predict the capacity degradation of LIB more accurately.