Deep Learning Methods for Multiscale Mechanical Problems

Intrinsic multiscale features in various physical systems stem from hierarchical structures that span a broad spectrum of temporal and spatial scales beyond the reach of any single modeling and simulation method, which have been recognized as significant challenges in multiscale engineering problems. Recent advancements in deep learning, particularly in deep neural networks, have shown remarkable successes in different scientific research fields. In this seminar, I will introduce three deep learning approaches we developed recently to address the multiscale challenge in different mechanical engineering problems, including a multiscale neural operator network model designed for inferring bubble growth dynamics across scales from nanoscale to macroscopic continuum scale, a neural operator model that functions as an effective surrogate for physics-based finite element models to predict the transient response of composite materials under dynamic loading conditions, and a deep learning framework integrated with physics-informed symbolic regression for model discovery directly from observational data. 

Dr. Zhen Li is Assistant Professor of Mechanical Engineering at Clemson University. He received his BS in Engineering Mechanics from Wuhan University in 2005 and PhD in Fluid Mechanics from Shanghai University in 2012. After a short postdoctoral experience in University of California-Merced, he joined the CRUNCH group at Brown University in 2013 as a postdoc and then was promoted to research assistant professor in 2016 and research associate professor in 2019. His research focuses on multiscale modeling for tackling challenges in multiscale/multi-physics problems, in particular on mathematical and physical foundation of scale-bridging, multiscale modeling methods and machine-learning approaches for complex fluids and soft matter, which are supported by DOE, NSF, NASA and ARO projects. Dr. Li has published more than 60 journal papers with an H-index of 26, and released multiple open-source software packages, including the Multiscale Universal Interface (MUI) library, the USERMESO GPU-accelerated particle simulator and the DPD-MESO module of LAMMPS.