Machine Learning-Driven Prediction of Elastic Modulus in Clay-Reinforced Polymer Nanocomposites


Nur Yılmaz G., Kandemir A. Ç.

MecaNano 2nd General Meeting , Vienna, Austria, 1 - 03 May 2024, pp.78

  • Publication Type: Conference Paper / Summary Text
  • City: Vienna
  • Country: Austria
  • Page Numbers: pp.78
  • TED University Affiliated: Yes

Abstract

Clay-reinforced polymer nanocomposites, particularly those incorporating Montmorillonitebased clays, have garnered significant attention for their potential to enhance mechanical properties in various engineering domains. These nanocomposites are pivotal in the automotive industry, aerospace, civil engineering, biomedical devices, and electronics, where material robustness and lightweight characteristics are crucial. Even though, state of the art includes numerous experimental analyses of various clay-reinforced nanocomposites, a comprehensive, data-driven analysis focusing on the elastic modulus of these nanocomposites, particularly from a multi-scale perspective of their mechanical behavior, remains largely unexplored. To assist nanomechanical researchers in this respect, this study focuses on harnessing machine learning techniques to predict the elastic modulus of polymer nanocomposites embedded with clay nanoparticles through the classification of key factors. By avoiding the need for extensive experimental procedures, this approach aims to streamline the material development process. Our methodology involves a comprehensive analysis of state-of-the-art studies. Key factors in our predictive model include the polymer matrix elastic modulus values, the size of Montmorillonite silicate layers, and the concentration of clay within the nanocomposite. Through this research, we aim to provide a classification-based machine learning approach for engineers and researchers to anticipate the mechanical performance of clay-reinforced polymer nanocomposites, thereby facilitating more efficient material selection and design in a wide range of applications