Figure 5 - uploaded by Peter Wellens
Content may be subject to copyright.
Free surface of a regular wave (period T = 6 s , height H = 4 m , length λ ≈ 50 m , water depth d = 10 m ) at t = 144 s (2 wave lengths). Simulations on different grids are compared to the analytical RF solution. Indicated grids cover the entire domain.
Source publication
The CFD simulation tool ComFLOW, developed for the simulation of slosh-ing liquids and two-phase flow, is applied to study water waves around a semi-submersible model. ComFLOW solves the Navier-Stokes equations in both water and compressible air, with second order accuracy in both space and time. The water surface is advected by means of a modified...
Context in source publication
Similar publications
This paper presents a study of the turbulent, compressible air flow induced by a fan impeller without scroll casing. The equations governing the flow – continuity equation, momentum equations, energy equation and the equation of state – were solved numerically using the finite volume method. The Reynolds Stress Model was used to provide the closure...
This paper presents a study of the turbulent, compressible air flow induced by an impeller of a high pressure fan. The equations governing the flow – continuity equation, Reynolds-averaged Navier-Stokes equations, energy equation and the equation of state – were solved using the finite volume method. The Reynolds stress model was used to provide th...
This paper analyzes the long-term response of unlined energy storage located at shallow depth to improve the distance between a wind farm and storage. The numerical approach follows the hybrid scheme that combined a mechanical constitutive model to extract stress and strains at the first cycle and polynomial-type strain accumulation functions to tr...
Citations
... To prevent from isolated droplets and mass losses, a local height function (LHF) has been introduced in ComFLOW , Veldman 2006, Kleefsman et al. 2004, 2005a, 2005b, Luppes et al. 2005, 2006, 2009b, 2010a, 2010b, 2011. The LHF is applied in a block of cells surrounding a central S-cell. ...
... At the same time, regular or irregular waves can enter the flow domain as prescribed. In this way, less grid-points are required compared to traditional damping-zone techniques, and hence computing times are reduced considerably (Luppes et al. 2009a, 2010a). ...
... A powerful Krylov method is needed, with incomplete LU-preconditioning for acceleration; the price to reduce spurious reflections. The performance of the ABC can be investigated , Luppes et al. 2009a, 2010a) by comparing the free surface at a measurement position (x m =200m) in two simulations, with either ABC or Sommerfeld at the outflow (located at x out =400m), see Fig. 7. As a reference, also a very long domain (x end = 2000m) is considered, where the surface elevations at x m cannot be disturbed by reflections from the outflow at x end . ...
The CFD simulation tool ComFLOW is developed for simulations of two-phase flow and wave impact in offshore and coastal applications. ComFLOW solves the Navier-Stokes equations in both water and air, with second order accuracy. The water surface is advected by means of an improved VOF method, through a local height function approach. Numerical reflections and spurious velocities are prevented by absorbing boundary conditions and gravity-consistent density averaging. The present focus lies on accurate wave propagation, the effect of viscosity in shear layers, regularisation models for turbulence, coupling algorithms for interactive body motion and enhanced numerical efficiency through local-grid-refinement and parallelisation. Copyright © 2013 by the International Society of Offshore and Polar Engineers (ISOPE).
... However, it is not adapted to deal with most coastal engineering simulations as it lacks specific boundary conditions for realistic wave generation and active absorption (Lara et al., 2011) or porous media as other 3D models recently available and del Jesus et al. (2012). Other models available have been used for similar purposes, for example ComFLOW®, which offers two phase flow simulations of waves with active wave absorption (Luppes et al., 2010). Nevertheless, its computational scheme is less robust, because it relies in the cutting cell method, which can have less computational cost, but also involves a less realistic behaviour compared with the adaptative meshing method of OpenFOAM®. ...
... There are several studies with different absorption theories, e.g., Christensen and Frigaard (1994), Schäffer and Klopman (2000) or Newman (2010). It has also been applied to numerical models as in Wellens et al. (2009), Luppes et al. (2010), Lara et al. (2011) or Wellens (2012). Wave generation in numerical models requires the use of special boundary conditions linked with active absorption to study the processes correctly, as it prevents from effects such as the increment of energy in the domain and the rising of the mean water level. ...
The present paper and its companion (Higuera et al., 2012) introduce OpenFOAM® as a tool to consider for coastal engineering applications as it solves 3D domains and considers two-phase flow. In this first paper, OpenFOAM® utilities are presented and the free surface flow solvers are analysed. The lack of specific boundary conditions for realistic wave generation is overcome with their implementation combined with active wave absorption. Wave generation includes all the widely used theories plus specific piston-type wavemaker replication. Also standalone active wave absorption implementation is explained for several formulations, all of which are applicable to 3D cases. Active wave absorption is found to enhance stability by decreasing the energy of the system and to correct the increasing water level on long simulations. Furthermore, it is advantageous with respect to dissipation zones such as sponge layers, as it does not increase the computational domain. The results vary depending on the theory (2D, Quasi-3D and 3D) but overall performance of the implemented methods is very good. The simulations and results of the present paper are purely theoretical. Comparisons with laboratory data are presented in the second paper (Higuera et al., 2012).
... However, the size of the domain is increased by some fraction of the wavelength, which adds significant computational cost to the case. The other approach is to use Dirichlet-type boundary conditions to generate and absorb the waves, as in the works presented by Luppes et al. (2010 and Higuera et al. (2013a), which does not noticeably alter the computational cost. ...
This paper and its companion Higuera et al. (2014--this issue) introduce the formulation of Volume-Averaged Reynolds-Averaged Navier–Stokes (VARANS) equations in OpenFOAM® to simulate two-phase flow through porous media. This new implementation, so-called IHFOAM, corrects the limitations of the original OpenFOAM® code. An innovative hybrid methodology (2D–3D) is presented to optimize the simulation time needed to assess the three-dimensional effects of wave interaction with coastal structures. The combined use of a 2D and a 3D model enables the practical application of the 3D VARANS code to simulate real cases, contributing to a significant speed-up. This is highly convenient and especially suitable for non-conventional structures, as it overcomes the limitations inherent to applying semi-empirical formulations out of their range or 2D simulations only. A detailed study of stability and overtopping for a 3D porous high-mound breakwater at prototype scale subjected to oblique irregular (random) waves is carried out. Pressure around the caissons, overtopping discharge rate and turbulent magnitudes are presented in three dimensions. The mean pressure laws present a high degree of accordance with the formulation provided by Goda–Takahashi. Furthermore, local effects due to three-dimensional processes play a significant role, especially close to the breakwater head.
... Therefore, RANS codes have already been able to deal with a great number of applications. A brief list of such cases includes all kinds of wave generation and absorption (Higuera et al., 2013a;Jacobsen et al., 2012;Lara et al., 2011;Lin and Liu, 1999;Troch and De Rouck, 1999) and wave interaction with coastal structures Guanche et al., 2009;Higuera et al., 2013b;Lara et al., 2006;Lara et al., 2008;Lara et al., 2012;Losada et al., 2008;Luppes et al., 2010). ...
In this paper and its companion (Higuera et al., 2014--this issue), the latest advancements regarding Volume-averaged Reynolds-averaged Navier–Stokes (VARANS) are developed in OpenFOAM® and applied. A new solver, called IHFOAM, is programmed to overcome the limitations and errors in the original OpenFOAM® code, having a rigorous implementation of the equations. Turbulence modelling is also addressed for k-ϵ and k-ω SST models within the porous media. The numerical model is validated for a wide range of cases including a dam break and wave interaction with porous structures both in two and three dimensions. In the second part of this paper the model is applied to simulate wave interaction with a real structure, using an innovative hybrid (2D–3D) methodology.
... Because of the e −kz -behaviour of Φ in z-direction, i.e. k 2 Φ out = ∂ 2 ∂z 2 Φ out , k is replaced by 2nd-order derivatives along the boundary. This leads to the ABC The numerical implementation of (13) at an outflow boundary, which coincides with the location of the horizontal velocity u, is described in [18,20,32]. In short, first the derivatives of the potential in (13) are formally replaced through ∂ ∂x Φ = u and ∂ ∂t Φ = −p−gz (Bernoulli equation). ...
... In [18,20,32] the performance of the ABC is investigated by comparing the free surface at a measurement position (x m = 200m) in two simulations, with ABC or Sommerfeld at the outflow (x e = 400m), see Fig. 8(right). As a reference, also a very long domain (x E = 2000m) is considered, where the surface elevation at x m cannot be disturbed by reflections from the outflow. ...
... This however significantly adds to the computational effort. The ABC clearly outperforms pressure-damping methods, since both reflections and computing times are considerably smaller [18,20,32]. ...
... However, the size of the domain is increased by some fraction of the wavelength, which adds significant computational cost to the case. The other approach is to use Dirichlet-type boundary conditions to generate and absorb the waves, as in the works presented by Luppes et al. (2010), del Jesus et al. (2012) and Higuera et al. (2013a), which does not noticeably alter the computational cost. Furthermore, 3D RANS codes have already been used to analyse a small number of processes and typologies providing accurate solutions (Higuera et al., 2013b; Lara et al., 2012). ...
This paper presents the validation results for the IH-3VOF model presented in Part I (del Jesus et al., 2011) of this paper. Initially the validation procedure is focused on wave-breaking conditions. A dam-break interacting with a prism is considered as a simplified approach to the study and analysis of wave-breaking conditions. Two different turbulence models are used in order to investigate flow separation prediction and its influence on the net forces exerted on the prism. Single and two-phase flow simulations are carried out to investigate the importance of trapped air within the fluid in the development of the forces acting on the prism. In addition, both a porous and a solid prism are considered for the study. Surface gravity waves interacting with a vertical structure in a narrow flume are presented next. Comparisons between experimentally and numerically obtained free surface elevation, pressure distribution and velocity around the structure are carried out. The hydrodynamics around the structure are studied in detail and compared with the numerical model predictions. Finally, the interaction of regular and solitary waves with a vertical breakwater in a wave basin is studied. The comparison between the laboratory tests and the numerical simulations are carried out for free surface elevation and pressure distribution over the structure under solitary and regular waves. The different results show a good performance of the IH-3VOF model, reproducing the most important processes that appear in the interaction of surface gravity waves with porous structures; therefore validating the new equations and the numerical implementation of the model.
... Therefore, RANS codes have already been able to deal with a great number of applications. A brief list of such cases includes all kinds of wave generation and absorption (Higuera et al., 2013a; Jacobsen et al., 2012; Lara et al., 2011; Lin and Liu, 1999; Troch and De Rouck, 1999) and wave interaction with coastal structures (del Jesus et al., 2012; Guanche et al., 2009; Higuera et al., 2013b; Lara et al., 2006; Lara et al., 2008; Lara et al., 2012; Losada et al., 2008; Luppes et al., 2010). One of the strong points of RANS is that they are accessible to the whole community through commercial codes, but also free and open source models are available. ...
... One of the strong points of RANS is that they are accessible to the whole community through commercial codes, but also free and open source models are available. Some examples of CFD codes applied to coastal engineering include IH-2VOF Lara et al. (2006), IH-3VOF Lara et al. (2012), COMFLOW Luppes et al. (2010), VOFbreak Troch and De Rouck (1999) or OpenFOAM ® Higuera et al. (2013b). However, to the authors' knowledge and until this work, there is no three-dimensional open source model available in which porous media flow is treated for two-phase flows. ...
This paper and its companion paper (Lara et al. (2012)) describe the capability of a new model, called IH-3VOF, to simulate wave–structure interaction problems using a three-dimensional approach, when porous structures are present. The lack of a universal approach for the formulation of porous media flow equations has motivated a new derivation in the present work. Applications dealing with heterogeneous media, where porosity varies along the porous body, such as the study of multilayered rubble-mound breakwaters, are the final objective of the study. In this first paper, a new derivation of the equations, eliminating the limitations imposed by previous approaches is presented. The model integrates a new set of equations which covers physical processes associated with flow interaction with porous structures. The model considers the multiphase VARANS equations, a volume-averaged version of the traditional RANS (Reynolds-Averaged Navier–Stokes) equations. Turbulence is modeled using a k–ε approach, not only at the clear fluid region but also inside the porous media. A VOF technique is used to track the free surface. In this first paper, the model has been validated using laboratory data of a two-dimensional flow. In the companion paper the model is further validated with new experimental data sets, considering porous and solid structures as well as the presence of air. The model predictions present an excellent agreement with the laboratory measurements.
... Because of the e −kz -behaviour of Φ in z-direction, i.e. k 2 Φ out = ∂ 2 ∂z 2 Φ out , k is replaced by 2nd-order derivatives along the boundary. This leads to the ABC The numerical implementation of (13) at an outflow boundary, which coincides with the location of the horizontal velocity u, is described in [18,20,32]. In short, first the derivatives of the potential in (13) are formally replaced through ∂ ∂x Φ = u and ∂ ∂t Φ = −p−gz (Bernoulli equation). ...
... In [18,20,32] the performance of the ABC is investigated by comparing the free surface at a measurement position (x m = 200m) in two simulations, with ABC or Sommerfeld at the outflow (x e = 400m), see Fig. 8(right). As a reference, also a very long domain (x E = 2000m) is considered, where the surface elevation at x m cannot be disturbed by reflections from the outflow. ...
... This however significantly adds to the computational effort. The ABC clearly outperforms pressure-damping methods, since both reflections and computing times are considerably smaller [18,20,32]. ...
The CFD simulation tool COMFLOW is developed for the simulation of sloshing liquids and two-phase flow in e.g. offshore applications. COMFLOW solves the Navier-Stokes equations in both water and air, with second order accuracy in both space and time. The water surface is advected by means of a modified VOF method, with improved accuracy through a local-height-function (LHF) approach. Numerical reflections are prevented by specially designed absorbing boundary conditions (ABC). Gravity-consistent density averaging for twophase flow prevents spurious velocities near the free surface. Several aspects in the numerical model in COMFLOW need further extension and improvement. In present research, the focus lies on accurate wave propagation and the effect of viscosity in shear layers (model small-scale flow details). The numerical efficiency is improved by speed-up through local-grid-refinement techniques and parallelisation. Other scientific items that receive attention are multi-dimensional non-reflecting boundary conditions and accurate turbulence modelling on coarse grids with regularisation models.
... Hence, a better understanding of wave impact forces is urgently needed. Roel Luppes, Bulent Duz, Henri J.L. van der Heiden, Peter van der Plas and Arthur E.P. Veldman The CFD simulation tool ComFLOW has been developed for the accurate prediction of hydrodynamic wave loading on offshore and coastal protection structures, up to a detailed level [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] . For example, maximum pressure forces, duration of pressure peaks and shear stresses can be computed everywhere on a structure. ...
... At the same time, regular or irregular waves can enter the flow domain as prescribed. Compared to traditional damping-zone techniques, less grid-points are required and computing times are reduced considerably [10][11][12] . ...
... With (2), reflection coefficients are significantly smaller over a wide range of the dimensionless wave number kh, than when Sommerfeld is used as boundary condition [10][11][12] . ...
The simulation tool ComFLOW, developed for accurate predictions of hydrodynamic
wave loading, is based on the Navier-Stokes equations. The free-surface dynamics
is described through VOF, with local height-function approach to enhance accuracy. Numerical
reflections are prevented by specially designed absorbing boundary conditions.
Gravity-consistent density averaging for two-phase flow prevents spurious velocities near
the free surface. Good progress has been made in the prediction of sloshing, green water
loading and wave run-up. In the ComFLOW3 project the focus is on wave propagation,
the effect of viscosity in shear layers (regularisation turbulence modelling), interactive
vessel-wave dynamics and improved numerical efficiency through local grid refinement.
... However, it is not adapted to deal with most coastal engineering simulations as it lacks specific boundary conditions for realistic wave generation and active absorption (Lara et al., 2011) or porous media as other 3D models recently available and del Jesus et al. (2012). Other models available have been used for similar purposes, for example ComFLOW®, which offers two phase flow simulations of waves with active wave absorption (Luppes et al., 2010). Nevertheless, its computational scheme is less robust, because it relies in the cutting cell method, which can have less computational cost, but also involves a less realistic behaviour compared with the adaptative meshing method of OpenFOAM®. ...
... There are several studies with different absorption theories, e.g., Christensen and Frigaard (1994), Schäffer and Klopman (2000) or Newman (2010). It has also been applied to numerical models as in Wellens et al. (2009), Luppes et al. (2010), Lara et al. (2011) or Wellens (2012). Wave generation in numerical models requires the use of special boundary conditions linked with active absorption to study the processes correctly, as it prevents from effects such as the increment of energy in the domain and the rising of the mean water level. ...
