RANS Simulations of Ship Motions in Regular and Irregular Head Seasusing the SWENSE Method

C. Monroy, G. Ducrozet, P. Roux de Reilhac, L. Gentaz, P. Ferrant, B. Alessandrini

Laboratoire de Mecanique des Fluides/EHGO (UMR CNRS 6598) Ecole Centrale de Nantes

Nantes, France

ABSTRACT

This paper presents recent advances of the SWENSE (Spectral WaveExplicit Navier-Stokes Equations) approach, a method for simulatingfully nonlinear wave-body interactions including viscous effects. Po-tential flow theory is used to compute the incident waves while viscouseffects are taken into account by using a Reynolds Averaged Navier-Stokes Equations (RANSE) solver to obtain the diffracted field in thefull domain. Arbitrary incident wave systems can be described, includ-ing regular, irregular waves, multidirectional waves and focused waveevents. The model may be fixed or moving with arbitrary speed and 6degrees of freedom motion.This paper is devoted to the validation of the SWENSE methodin irregular head waves, in the case of an extensively long 2 DoFsimulation. Results of the present approach compare favorably withexperimental data.

KEY WORDS: RANS Equations; potential flow; nonlinear flow;combined approach; wave-body interactions; HOS model; SWENSEmethod.

INTRODUCTION

Performance and seakeeping predictions are usually carried out in tow-ing tank. However, in ship hydrodynamics, Computational Fluid Dy-namics (CFD) is more and more used as a practical design tool. Mainadvantages of CFD are cost and time reduction as well as easier ac-cess to detailed flow field information. The complexity of simulatingthe behaviour of a ship in seaways was historically overcome by sep-arating the problem in many simpler analysis: resistance, propulsion,maneuvering and seakeeping. Although these aspects are strongly cou-pled, CFD tends to simulate each of these phenomena separately usingadapted theories :

• Resistance and propulsion analysis are now often addressed usingviscous flow solvers based on the solution of RANS Equations,because viscosity or flow separation effects play an important rolein the physics of those phenomena.

• Maneuvering and seakeeping problems are still currently solvedby potential flow theory which is less time consuming and enablesan accurate and efficient account of wave propagation phenomena.

However neglecting viscous effects can lead to poor predictions espe-cially for rolling motion or cases for which strong separation occurs.This is why a natural evolution for CFD is to try to address seakeep-ing and resistance problems within a unified approach by taking intoaccount incident waves in performance predictions.The classical method used to simulate the viscous flow around a shipadvancing in head waves is to impose an incident wave field at the inletboundary. It is modelled as velocity and pressure perturbations whichare added to the uniform stream. These perturbations are usually derivedfrom the linear potential flow solution for free-surface travelling waves.However such simulations require very large computing resources be-cause grids must be very refined between the location of the structureand the location of the wave generation systems. This is indeed neces-sary to propagate waves from the paddle to the structure with no notice-able damping. Moreover successive wave reflections on the body or onthe paddle affect the incoming wave train and reduce the useable dura-tion of the numerical simulation ; it is indeed very complicated to dampthe diffracted field without modifying the incident waves.Yet, to be fair, it must be said that RANSE seakeeping simulations ofa ship advancing in head regular waves using this straightforward ap-proach are realizable and have been presented in Weymouth et al. (2005)or more recently in Visionneau et al. (2008) showing good results com-pared to ”state of the art” potential simulations. However, the generationof irregular wave trains or focused waves will be very problematic withthis method, especially for 3D sea states.

To overcome these difficulties an original formulation is used here bymodifying the initial problem in order to solve the diffracted flow only.This approach has previously been used in the frame of potential theory,by Di Mascio et al. (1994) or Ferrant (1996) in 3D cases. It consists insplitting all unknowns of the problem (potential and free-surface eleva-tion) into the sum of an incident term and a diffracted term. The incidentterms are described explicitly using a nonlinear potential flow model.Thus only the part of the grid in the vicinity of the structure needs to berefined. Far from the body a stretched grid allows an efficient dampingof the diffracted flow.

Proceedings of the Nineteenth (2009) International Offshore and Polar Engineering ConferenceOsaka, Japan, June 21-26, 2009Copyright © 2009 by The International Society of Offshore and Polar Engineers (ISOPE)ISBN 978-1-880653-53-1 (Set); ISSN 1098-618

458