Interpretable Augmented Intelligence for Multiscale Material Design
This collaborative project aims at building an effective and interpretable learning framework for material data across scales to solve a major challenge in current data-driven materials design. Our test systems are composites and alloys.
Expertise in material science and data science will be leveraged to develop and apply interpretable and physics-informed data science methods for revolutionary understanding of mechanical properties of polymer composites and alloys using large sets of 3D computational data, images, and tomography. We will utilize available data more efficiently through combination with physical rules and prior knowledge to develop an interpretable augmented intelligent framework to learn principles behind the association of input structures with material properties with uncertainty quantification. The interconnected tasks involve the (1) collection and curation of large amounts of computational and experimental data for polymer/carbon nanotube composites and alloys from open data sources and targeted calculations and experiments, (2) the development of geometric and topological methods incorporating physical principles to generate a better, more sensitive low-dimensional representation of the multidimensional data and characterize the parameter space related to mechanical properties, (3) the development of a Bayesian deep reinforcement learning framework to generate interpretable knowledge graphs that depict the relational knowledge among physical quantities with uncertainty quantification, and (4) the prediction of mechanical properties to reveal design principles to improve materials performance, evaluate and validate the proposed methods, and develop software for dissemination. The methods and results will accelerate the development of ultrahigh strength and lightweight carbon-based composites for aerospace applications and multi-element alloys for more durable engine parts by guiding in the large possible design space and providing faster predictions than experiments and traditional simulation methods. New methods and computational algorithms will become publicly available. The investigators will train PhD students and undergraduate students from various disciplines with a focus on engaging women and minorities in STEM fields and develop short courses that integrate novel material science applications and data science methods.
We are grateful for support from the National Science Foundation (CMMI 1940335).