Design of polymers for energy storage capacitors using machine learning and evolutionary algorithms

To meet the demands of emerging electrification technologies, polymers that are capable withstanding high electric fields at temperatures needed. Given staggeringly large search space polymers, traditional, intuition- and experience-based Edisonian approaches too slow discovering new can these demands. In this work, a genetic algorithm was combined with five machine learning-based property predictors to design over 50,000 hypothetical achieve target properties. Additionally, polymer synthesizability-based criterion used narrow down 23 candidates likely be synthesizable 3665 may synthesizable. A version code is also made available for public use on GitHub.