EXTENDED ABSTRACT: The automotive industry is focusing on developing lightweight and eco-friendly vehicles in response to climate change, with active research and development underway to enhance the power efficiency of electric and hybrid vehicles. In this context, efforts are being made to develop polymer composites with excellent thermal conductivity that meet mechanical property requirements, as well as to integrate them into components. In this presentation we propose a method that utilizes raw material information and product information of various polymers to develop a machine learning model to predict the mechanical and thermal properties of polymer composites. We find that integrating representations of product and ratio information shows a good predictive performance and significantly enhance the model accuracy. This research may offer valuable insights for data-driven approaches to build accurate prediction models.
Keywords: thermal conductive polymer composites, machine learning, property prediction, data representation
In Kim is a Senior Researcher at the Korea Research Institute of Chemical Technology (KRICT) in the Chemical Materials Solutions Center from 2020. He earned his Ph.D. in Graduate School of EEWS (Energy, Environment, Water, and Sustainability) from Korea Advanced Institute of Science and Technology in 2018 under the supervision of Prof. Yousung Jung. His research focuses on the development of artificial intelligence platforms to accelerate the design of polymer composites and soft materials, as well as multiscale modeling and simulations of organic and semiconductor materials. In has co-authored numerous publications in top journals, contributing to advancements in materials science and technology.