State of Health Estimation of Lithium Ion Battery Based on CNN-LSTM Neural Network

Abstract

With the development of new energy, lithium-ion batteries are widely used in electric vehicles and energy storage. Lithium-ion battery health status is the key technology of battery management system. Accurate estimation of battery health state is the key to ensure the safe and stable operation of batteries. In this paper, three factors with a high correlation with the state of health are proposed as battery external health features, and a data-driven CNN-LSTM neural network prediction method is constructed. By NASA’s battery data sets, the method is proved by the experimental results show that this method can more accurately predict the health status of lithium-ion batteries.