Lori Graham-Brady, Ph.D.
Professor of Civil and Systems Engineering
Johns Hopkins University
Friday, Feb 25th at 3pm
Sid & Marian Green Classroom (3550 MEK)
ABSTRACT: Materials characterization and/or the significant effort associated with computational mechanics can lead to a sparsity of data that makes proper uncertainty quantification and/or multi-scale modeling extremely challenging. Machine learning (ML) provides one path to efficient surrogates. ML has the added advantage of supporting digital reconstruction of materials that allow for rapidly generated virtual microstructures that are statistically representative of the actual material. Recently developed approaches allow digital reconstruction of three-dimensional stochastic microstructures based on transfer learning, which provide a statistically significant set of digitally generated samples. Such samples enable assessment of the uncertainty associated with random microstructure sampling. In the context of multi-scale modeling, one can use this digital generation process to create virtual microstructures that represent key integration points. Furthermore, an encoder-decoder Unet architecture is trained to perform an image to image mapping from composite microstructure to stress distribution, providing a rapid approach for identifying micro-scale stresses from these virtual microstructures. This talk will focus on these two approaches, discussing some of the very promising results, their potential to enhance uncertainty quantification and/or multi-scale models for mechanics of multi-phase materials. The potential limitations of the proposed techniques that can be addressed in the future will also be discussed.
BIO: Lori Graham-Brady is Professor of Civil and Systems Engineering at Johns Hopkins University, with secondary appointments in Mechanical Engineering and Materials Science & Engineering. She served as Department Chair from 2015-2021. Her research interests are in computational stochastic mechanics, multiscale modeling of materials with random microstructure and the mechanics of failure under high-rate loading. She is the Director of the ARL-funded Center for Materials in Extreme Dynamic Environments and Associate Director of the Hopkins Extreme Materials Institute. She has received a number of awards, including the Presidential Early Career Award for Scientists and Engineers (PECASE), the Walter L. Huber Civil Engineering Research Prize, and the William H. Huggins Award for Excellence in Teaching.