A review of the recent progress in battery informatics

Abstract

Batteries are of paramount importance for the energy storage, consumption, and transportation in the current and future society. Recently machine learning (ML) has demonstrated success for improving lithium-ion technologies and beyond. This in-depth review aims to provide state-of-art achievements in the interdisciplinary field of ML and battery research and engineering, the battery informatics. We highlight a crucial hurdle in battery informatics, the availability of battery data, and explain the mitigation of the data scarcity challenge with a detailed review of recent achievements. This review is concluded with a perspective in this new but exciting field.

Descriptor-Free Design of Multicomponent Catalysts

“…Encoding both the accuracy of prediction and the uncertainty of prediction into the searching strategy also alleviate the requirement of data amount for the optimization relying on accurate prediction only, which is usually a necessary condition in other material informatics approaches. 44,63 To understand the switch between exploration and exploitation in the current study, Figure 4c shows the uncertainty against the predicted activity for every recommended composition. Due to the dominance of compounds with low-to-moderate activities in the initial set, the search in the first three iterations was characterized by relatively low predicted mean activity.…”

“…Herein, we report a Bayesian optimization (BO) approach that directly connects a ML-guided screening to the experimental synthesis and evaluation of multicomponent catalysts without the aid of physical descriptors (Figure b). BO is an effective strategy to integrate the modeling and acquisition of prediction with the measurement and validation of prediction for accelerated global optimization. , For research of multicomponent catalysts, BO has been used to screen high entropy alloy catalysts for oxygen reduction from computational data and mesoporous Pt x Pd y Au 1– x – y films for methanol oxidation (an optimization of two independent variables) . The current work employed BO in the much more complicated task of optimizing a seven-component (six independent variables) oxide catalyst for the direct decomposition of nitric oxide.…”

Na superionic conductor-type LiZr2(PO4)3 as a promising solid electrolyte for use in all-solid-state Li metal batteries

“…Various informatics-aided approaches are proposed for the prediction and/or optimization of battery materials. 83,84 For example, Joshi et al predicted the voltage of electrode materials by DFT calculations by machine learning. 85 More recently, Louis et al suggested deep neural network based approach for a prediction of voltage of battery electrode.…”

Informing geometric deep learning with electronic interactions to accelerate quantum chemistry

“…Discovering new molecules and materials is central to tackling contemporary challenges in energy storage and drug discovery (1,2). As the experimentally uninvestigated chemical space for these applications is immense, large-scale computational design and screening for new molecule candidates have the potential to vastly reduce the burden of laborious experiments and to accelerate discovery (3)(4)(5).…”