Disputation: Tormod Ravnanger Landet
Doctoral candidate Tormod Ravnanger Landet at the Department of Mathematics, Faculty of Mathematics and Natural Sciences, is defending the thesis "Discontinuous Galerkin methods for multiphase flow" for the degree of Philosophiae Doctor.
Tormod Ravnanger Landet
The University of Oslo is closed. The PhD defence and trial lecture will therefore be fully digital and streamed directly using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.
Ex auditorio questions: the chair of the defence will invite the audience to ask ex auditorio questions either written or oral. This can be requested by clicking 'Participants -> Raise hand'.
Join the disputation (deactivated)The meeting opens for participation just before 10am, and closes for new participants approximately 15 minutes after the defense has begun.
Submit the request to get access to the thesis.
"The role of CFD in Marine design. What are the pros and cons of CFD versus physical model tests. Where are they superior, respectively, and how can one utilize them combined?"
Main research findings
A new method for combined air and water simulation
Accurate simulation of water waves is a crucial part of designing safe and energy efficient ships, and for developing new uses of the ocean such as floating windmills. Current simulation methods are most often based on dividing the physical world into small boxes, and then keeping track of average values—such as the average velocity—in each box. More advanced methods exist, where each box contains not only information about the average velocity, but also on how the velocity changes within each box. Such methods are, however, very sensitive to non-smooth transitions—such as the sharp jump between water and air—and will hence most often give non-physical (wrong) results without proper stabilisation.
The computer will sometimes need to use very small boxes in order to describe the geometry of a complex surface, such as the air/water surface of a breaking wave. In such cases the existing methods can often give sufficiently accurate descriptions of the physics. Away from the areas where the geometry changes rapidly it is, however, quite wasteful to let the computer keep track of thousands of small boxes containing average values, when only a few more advanced (higher order) boxes containing information on the smooth change inside each box would be sufficient.
My PhD research has resulted in a computational method that uses higher order boxes for describing the air and water physics. The new method includes a specially tailored stabilisation strategy that prevents the simulations from giving non-physical results when the fluid density abruptly changes from the light air to the heavy water at the water surface. By using high-order boxes (discontinuous Galerkin finite elements), the portion of the simulated world away from the surface can be accurately approximated with very few numerical boxes. The thesis contains some example simulations that show the same behaviour—both qualitatively and quantitatively—as published lab experiments, confirming that the method works. All the included simulations are reproducible, with the computer code available at ocellaris.org and all input and output files available on Zenodo.