Computational modeling for lightweight composite materials

Composite materials (CM) have become an ideal choice for automobiles and construction industries. The diversified applications of new CM demand an extensive investigation of their properties under elevated conditions. Nowadays, machine learning (ML) works as a powerful tool for data-driven modeling and gives an unprecedented insight to explore the properties of a system. New CM and their properties often confront uncertainties due to the chemistry of different matrices (polymer matrix, cement matrix, etc) and due to the variations in reinforcement (fiber reinforcement) and nanofillers (nanoparticles, nanotubes). Reduction in uncertainty, increment in production and controlling cost at the same time is a current challenge of ML. Therefore, the aim of the project is to develop a theoretical framework (novel optimized solution) with validation based on ML algorithms for uncertainty quantification, cost optimization, failure identification. This project will combine several ML algorithms to improve the properties of existing materials and proposing new materials with enhanced properties.