I am working with free surface flows (ocean waves) in the fluid mechanics research group. My research is on simulation of wave slamming impacts. I am working with a domain decomposition approach where a stochastic and non-linear incoming wave train is described by potential flow theory and the area around the studied structure is described by the Navier-Stokes equations with free surface capturing to handle slamming impacts, wave breaking and spray. The current plan is to use the discontinuous Galerkin method for the Navier-Stokes domain using the FEniCS software. The Laplace equation solver for the potential theory outer domain has not been decided on yet.
The goal is to be able to run long simulations of structures exposed to ocean waves. From these simulations slamming impact pressures and forces can be recorded. Given enough such impacts a statistical distribution of slamming loads can be established. Such a distribution can be used for determining the expected lifetime extreme load on a structure.
The studied structures are typically small equipment hanging of the side on a large ship, such as the life boats on the cruise ship above. It is difficult to a priori establish a critical wave slamming event that will cause the lifetime extreme load on the life boat due to the combined effect of radiated and diffracted waves that create strongly non-linear patterns along the ship side. Other structures hanging on the side of the ship may also disturb the flow. The second life boat from the front will have an inflow that is shaped by what happened when the wave hit the first life boat.
A coupled simulation of the non linear wave propagation shaped by the cruise ship's hull in an outer domain and the slamming on the life boats in an inner domain is one way to tackle this problem in a numerical simulation. The current industry standard for such slamming load assessments is to run extensive model test campaigns. Running long simulations where the whole cruise ship is inside the Navier-Stokes solver domain is currently not feasible.
In the inner domain free surface capturing is applied and the Navier-Stokes equation are solved. Keeping this inner domain to a minimum is vital to enable simulations of long enough duration to capture enough events for determining the statistical distribution of slamming loads. It is the goal of the project to give accurate load assessments without including the whole ship in the inner domain, only the immediate surroundings of the structure in question.
I am currently working on implementing the above in a program called Ocellaris. Click the link to see a description and the source code. The code is very much a work in progress and should not be assumed to work for any real problem.
I have a master of Marine Hydrodynamics from NTNU. After completing the master I worked for six years with ship hydrodynamics in the maritime advisory part of DNV GL. I am currently on leave from that position to work towards obtaining a PhD from UiO (2014-2018).