In ocean engineering, high-fidelity analyses like computational fluid dynamics Unsteady Reynolds Averaged Navier Stokes simulations are fundamental for the evaluation of realistic hydrodynamic loads during extreme events, which are usually overestimated by boundary element method based solvers, thus leading to the definition of an oversized wave energy converter structure. To describe multiphase flows, the most used model in RANS CFD simulations is Volume of Fluid model, which consists in creating an interface between the different fluids and assigning to each cell of the domain a percentage of each phase. Unfortunately, the model has a well-known problem in the ocean engineering community which increases with simulation time and affects the wave propagation. For the correct spatial propagation of the wave motion, it is necessary to use a very low Courant-Friedrich-Lewy factor, hence causing a noticeable increase in the computational time and speeding up the generation of artificial turbulence in the domain. The problem is intrinsic in the Volume of Fluid model and cannot be solved by refining the mesh or increasing the order of accuracy of the solvers. The numerical turbulence spreads in the whole domain and dampens the waves overtime after several wave periods (10-50). With this work, we analyse the effect of a threshold on the turbulent viscosity to reduce the production of numerical turbulent kinetic energy; theaim is to produce a general methodology to simulate long time frames in a RANS simulation with the Volume of Fluid model. The investigation is first performed in a semi-2D tank by monitoring the wave probes over time. It is then repeated in a 3D domainwith and without the limiter in the case of a floating body, to verify that the kinematics of the floater is not altered by the limiter.

Solutions to wave damping over time in CFD RANS simulations due to exponential generation of numerical turbulence / Casalone, P.; Dell'Edera, O.; Fontana, M.; Battisti, B.; Mattiazzo, G.. - 8:(2022). (Intervento presentato al convegno ASME 2022 41st International Conference on Ocean, Offshore and Arctic Engineering, OMAE 2022 tenutosi a deu nel 2022) [10.1115/OMAE2022-81447].

Solutions to wave damping over time in CFD RANS simulations due to exponential generation of numerical turbulence

Casalone P.;Dell'edera O.;Fontana M.;Battisti B.;Mattiazzo G.
2022

Abstract

In ocean engineering, high-fidelity analyses like computational fluid dynamics Unsteady Reynolds Averaged Navier Stokes simulations are fundamental for the evaluation of realistic hydrodynamic loads during extreme events, which are usually overestimated by boundary element method based solvers, thus leading to the definition of an oversized wave energy converter structure. To describe multiphase flows, the most used model in RANS CFD simulations is Volume of Fluid model, which consists in creating an interface between the different fluids and assigning to each cell of the domain a percentage of each phase. Unfortunately, the model has a well-known problem in the ocean engineering community which increases with simulation time and affects the wave propagation. For the correct spatial propagation of the wave motion, it is necessary to use a very low Courant-Friedrich-Lewy factor, hence causing a noticeable increase in the computational time and speeding up the generation of artificial turbulence in the domain. The problem is intrinsic in the Volume of Fluid model and cannot be solved by refining the mesh or increasing the order of accuracy of the solvers. The numerical turbulence spreads in the whole domain and dampens the waves overtime after several wave periods (10-50). With this work, we analyse the effect of a threshold on the turbulent viscosity to reduce the production of numerical turbulent kinetic energy; theaim is to produce a general methodology to simulate long time frames in a RANS simulation with the Volume of Fluid model. The investigation is first performed in a semi-2D tank by monitoring the wave probes over time. It is then repeated in a 3D domainwith and without the limiter in the case of a floating body, to verify that the kinematics of the floater is not altered by the limiter.
2022
978-0-7918-8593-2
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2981753