Multi-Physics and Multi-Scale Computations for Complex Problems in Engineering

In an era of ever-expanding computational capabilities, one might expect that large-scale, fully coupled simulations of multi-physics problems in science and engineering would have become commonplace, with abundant processor power enabling monolithic, simultaneous solution of multiple equation systems and domains. This has not been the case. More typically, increased capability is used to increase the fidelity of individual physics, with disparate physical phenomena solved separately and coupled through source or boundary condition terms. This approach to coupling has several important advantages compared with full monolithic solves, but if not applied carefully can lead to losses of accuracy, conservation, and stability. In this talk, two examples of coupling are discussed in detail. In the first, conjugate heat transfer, heat transfer between a fluid and solid, is examined as a representative example of multi-domain coupling. It is shown that common methods for advancing the solution in time are unstable for certain parameter choices, and an alternative approach is presented including a coupling parameter that can be optimally derived for stability and accuracy. In the second example, a new method is presented for coupling disparate models at course and fine scales; our method allows for conservation of energy in transient simulations. This approach is motivated by our group’s work on trying to predict material microstructure resulting from additive manufacturing processes in metals.

Bio: Greg Wagner received his Ph.D. in Mechanical Engineering from Northwestern University in 2001. He spent over 12 years as a staff member and later manager in the Thermal/Fluid Science and Engineering department at Sandia National Laboratories in Livermore, CA, where his work included multiscale and multiphysics computational methods, multiphase and particulate flow simulation, extended timescale methods for atomistic simulation, and large-scale engineering code development. In January 2015 he joined the faculty of the Mechanical Engineering department at Northwestern. His current research focuses on applying novel simulation methods and high performance computing to multiphase flows and flows with complex and moving geometries.