25 Years of Hydrodynamics CFD: success, impact of HPC, missing physics, and open-source software
Speaker: Eric Paterson, Virginia Polytechnic Institute and State University
Department: Mechanical & Aerospace Engineering
Location: Bowen Hall Auditorium 222
Date/Time: Friday, April 5, 2013, 3:30 p.m. - 4:30 p.m.
Over the past 25 years, the field of computational ship hydrodynamics has made extraordinary progress. Guided by international workshops and collaborative experiments, and supported by growth in high-performance computing (HPC), URANS and hybrid RANS/LES CFD is now capable of solving problems associated with model-scale resistance and propulsion, maneuvering, and seakeeping. However, many multi-physics problems remain unsolved, with discrepancy between theory, experiments, computer models, and full-scale observations. In this talk, I will give a brief overview of this history to motivate my thesis: a) that reductionism, while critical to the basis of engineering knowledge, often blinds us from the truth; b) that open-source computational mechanics software provides a pathway for incorporating missing physics; and c) that the phenomenal growth in HPC is a critical research tool and resource that must be intelligently utilized, and guarded against driving research in unwarranted directions. As an example, I will present the theory of modeling inhomogenous currents (jets and wakes) in the presence of ocean-surface waves. The model is analogous to the physics of Craik-Leibovich vortex forcing which has been attributed as the mechanism for explaining Langmuir circulations. The model has been implemented in CFD solver based upon the open-source library OpenFOAM. Simulation results will be presented for both surface and submerged jets in the presence of both short- and long-wavelength waves. It will be shown that surface waves have a large influence on the advection of the current, and in generation of large-scale circulations.