ICE Seminar Co-sponsored by MAE – Supercomputers and the Turbulence Problem

Details

INSTITUTE FOR COMPUTATIONAL
ENGINEERING
Co-Sponsored by MAE Department

Seminar Series Presents:
Prof. Perry Johnson
University of California, Irvine

Supercomputers and the turbulence problem

In his Lectures on Physics (in the early 1960s), Nobel Laureate Richard
Feynman described the mathematical analysis of fluid turbulence as “the
central problem we ought to solve someday.” Around the same time,
Gordon Moore and others began forecasting the sustained exponential
growth in computational technology that has since transformed every field
of science and engineering. In this talk, I will argue that successful
prediction and analysis of societally-relevant turbulent flows cannot rely
(now or in the foreseeable future) solely on a brute force numerical
discretization to the governing partial differential equations. On the other
hand, fundamental advances in basic physical and mathematical
understanding of turbulent motions are still required to stimulate and
inform the creation of effective lower-cost computational frameworks.
Synergistically, direct numerical simulations of turbulent flows, when
feasible, provide crucial data to support such advances. Two in-progress examples will be presented: (i) the cascade of kinetic energy from large to small scales, and (ii) the enhancement of skin friction via turbulent momentum flux across boundary layers. Perry Johnson earned his Ph.D. in 2017 from Johns Hopkins University (advisor: Charles Meneveau), where his work on velocity gradient dynamics in turbulence won the Corrsin-Kovasznay award. He was then a postdoctoral fellow at the enter
for Turbulence Research at Stanford University for three years, working on various topics related to small-scale turbulence, multiphase & particle-laden flows, and boundary layers. He joined the Mechanical and Aerospace Engineering department at the University of California, Irvine, in 2020 as an assistant professor. His recent
research on the energy cascade was featured in Physics Today, and his forthcoming review of multi-scale velocity gradient dynamics will appear in the next issue of the Annual Review of Fluid Mechanics.