Prof. Guglielmo Scovazzi
Department of Civil & Environmental Engineering
Duke University

Simulations of Complex-Geometry Systems: Challenges and Solutions in the Era of Digital Twins

ABSTRACT: Creating Digital Twins (DTs) of mechanical systems or architected materials usually requires automatically interfacing direct numerical methods with meta-algorithms from optimization, uncertainty quantification, reduced order modeling, machine learning, etc. Developing DTs is challenging when geometries are complex, as grid generation becomes time- and labor-intensive, especially when uncertain or imperfect scans of the real world replace Computer-Aided Design (CAD) geometric datasets as inputs. In some extreme cases, the limitations of grid generation make DTs unfeasible. Finite Element Methods (FEMs) with immersed boundaries seem appealing in such scenarios, as they simplify grid generation. Unfortunately, immersed FEMs involve complex cell-cutting operations at boundaries, which have negative consequences on the conditioning and stability of the ensuing algebraic problems. This talk introduces the “Shifted Boundary Method” (SBM), a stable and accurate immersed FEM that eliminates cell-cutting operations. Boundary conditions are imposed on a surrogate (approximate) boundary, composed of full grid edges and faces, where Taylor expansions mimic the effect of the nearby true boundary. The performance of the SBM is demonstrated in large-scale linear and nonlinear problems in fluid mechanics, solid mechanics, fracture mechanics, and porous media flows.

BIOGRAPHY: Guglielmo Scovazzi is a Professor of Civil & Environmental Engineering and Mechanical Engineering & Materials Science at Duke University. His interests lie in the general areas of scientific computing and computational mechanics, specifically in computational fluid dynamics (CFD), computational solid mechanics, fluid-structure interaction, computational geomechanics, and the simulation of porous media flow. He earned a Bachelor of Science and a Master of Science (M.S.) in Aerospace Engineering from Politecnico di Torino in 1998. He received an M.S. degree in 2001 and a Ph.D. in 2004, both from the Mechanical Engineering Department at Stanford University. Between 2004 and 2012, he held the position of Senior Member of the Technical Staff at Sandia National Laboratories in Albuquerque, New Mexico. He has been a member of the faculty in the Civil & Environmental Engineering Department at Duke University since 2012. Guglielmo Scovazzi is a recipient of the 2014 Early Career Award from the Office of Science of the U.S. Department of Energy (DOE) Advanced Scientific Computing Research (ASCR) program, as well as the 2017 Presidential Early Career Award for Scientists and Engineers (PECASE). In February 2018, he was named a Kavli Fellow by the National Academy of Sciences and the Kavli Foundation, acknowledging the contributions of U.S. scientists under the age of 45. He was also named Distinguished Adjunct Professor at Pohang University of Science and Technology (POSTECH) in South Korea. He is an Associate Editor of the Journal of Computational Physics (Elsevier) and Computational Geosciences (Springer), and serves on the editorial board of Computer Methods in Applied Mechanics and Engineering (Elsevier), the International Journal on Numerical Methods for Fluids (Wiley), Engineering with Computers (Springer), and Advances in Computational Science and Engineering (American Institute of Mathematical Sciences). He is a member of the International Association for Computational Mechanics (IACM), the Society for Industrial and Applied Mathematics (SIAM), the U.S. Association for Computational Mechanics (USACM), the American Society of Civil Engineers (ASCE), and the American Society of Mechanical Engineers (ASME).