Recently, our gourp, together with Tsinghua University and Ulster University in the UK, published a review article on the use of in-depth learning to predict the state of lithium ion batteries in the journal of energy chemistry (JCR Q1, IF=13.599). The first author is Kai Luo, a 2020 master's degree candidates. This paper reviews the current widely used equivalent circuit and electrochemical models for lithium-ion battery state estimation, introduces, compares and summarizes the commonly used shallow learning models, such as linear regression, support vector machine, decision tree, etc., and deep learning methods, including convolutional neural network Cyclic neural networks and their variants have made the latest progress in building data driven models to predict the health and state of charge of lithium ion batteries, and further emphasized the advantages and limitations of in-depth learning, as well as the measures that can be taken to address the current problem of scarce high quality data.

URL for this paper:

https://www.sciencedirect.com/science/article/abs/pii/S2095495622003564