Machine learning research advances in energy storage polymer-based dielectrics

Abstract

In the new circumstances of modern scientific research combining advanced analytics and artificial intelligence, the application of machine learning (ML) to energy storage dielectrics has become the focus of research attention in this field. In this review, the current disciplinary fields and basic workflow of ML applications are summarized and the important impact of ML in energy storage polymer-based dielectric research is emphasized, with a focus on enabling rapid performance prediction and accelerating the research and development of novel materials. The content focuses on several common methods and representative algorithms for establishing databases, including dataset collection results, model calculation results, and experimental verification results. Moreover, the advantages and disadvantages of each method of dataset collection and the accuracy and reliability of each algorithm prediction application are summarized and compared. Finally, based on ML’s impact on the research field of energy storage polymer, its prospects and challenges are discussed. This review not only provides the latest progress of existing researchers in using ML in energy storage polymers but also looks forward to providing new modes for the preparation of high-energy storage polymer-based dielectrics through ML.