Applying Machine Learning to Rechargeable Batteries: From the Microscale to the Macroscale

Emerging machine learning (ML) methods are widely applied in chemistry and materials science studies and have led to a focus on data-driven research. This Minireview summarizes the application of ML to rechargeable batteries, from the microscale to the macroscale. Specifically, ML offers a strategy to explore new functionals for density functional theory calculations and new potentials for molecular dynamics simulations, which are expected to significantly enhance the challenging descriptions of interfaces and amorphous structures. ML also possesses a great potential to mine and unveil valuable information from both experimental and theoretical datasets. A quantitative structure-function correlation can thus be established, which can be used to predict the ionic conductivity of solids as well as the battery lifespan. ML also exhibits great advantages in strategy optimization, such as fast-charge procedures. The future combination of multiscale simulations, experiments, and ML is also discussed and the role of humans in data-driven research is highlighted.

Automated summary

This paper summarizes the application of machine learning in rechargeable batteries, including exploring new functional theory calculations and molecular dynamics simulations, as well as mining valuable information from experimental and theoretical datasets. This has led to the establishment of a structure-function correlation for predicting ionic conductivity and battery lifespan, along with advantages in strategy optimization.