Data-driven approaches enabling the screening and design of promising materials for energy storage batteries
Data-driven approaches enabling the development of new materials are a research focus of the fourth research paradigm that promotes material innovation and accelerates the cross-integration of multidisciplinary and multifield material applications.As a crucial data-driven approach Integrating the first-principles method,Machine learning(ML)shows great potential to address the issues in interdisciplinary fields,such as materials science,chemistry,physics,and computer science,and elicits a promising avenue for the rapid development of new materials for energy storage batteries.To better understand this emerging field,this review systematically outlines the latest progress in high-throughput computational screening and ML that show great potential in the research of energy storage battery materials.Moreover,online material databases that are widely used in the discovery and design of materials are summarized,the construction of a new database is presented along with an example,and several issues encountered during data collection are discussed.Furthermore,this review discusses in detail the application examples associated with ML methods in terms of high-throughput computational screening,material property prediction,structure-electrochemical performance relationship,and material design.Finally,the challenges of ML in the field of energy storage batteries are analyzed and discussed.In addition,we provide perspectives on the applications of ML in energy storage materials.
万方数据
维普
