BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Direct Numerical Simulation of Liquid-Vapor Phase Change. Applicat
ions to Leidenfrost Droplet and Nucleate Boiling
DTSTART;VALUE=DATE-TIME:20161212T110000Z
DTEND;VALUE=DATE-TIME:20161212T113000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-569@indico.math.cnrs.fr
DESCRIPTION:Speakers: Sébastien Tanguy (Institut de Mécanique des Fluide
s de Toulouse)\nStudies on two-phase flows are of interest in many fundame
ntal problems and industrial\napplications\, as the spray formation in int
ernal combustion engine\, the bubble formation in heat\nexchangers\, the f
luid management in satellites or space launcher tanks\, the spray cooling
or the\ninteraction of bubbles with acoustic waves. The Direct Numerical S
imulation is a powerful tool\,\nwhich is complementary to experimental mea
surements\, to provide accurate results in complex\nsituations. However\,
unlike single-phase flows\, currently the direct numerical simulation of t
wo-\nphase flows cannot be considered as a fully mature field\, especially
in most configurations\ninvolving strong coupling between the interface m
otion with heat and mass transfer\, acoustic or\nshock waves\, and/or a so
lid boundary where a contact line can be formed. This presentation will\ne
mphasize on the development of new numerical methods [1\,2\,3\,4\,5] to pe
rform accurate Direct\nNumerical Simulations of two-phase flows with phase
change in the framework of sharp\ninterface capturing numerical methods.
The presentation will focus mainly on two specific\nconfigurations involvi
ng liquid vapor phase change\, i.e. Leidenfrost droplets and nucleate boil
ing.\nWe will discuss about suited numerical strategy to succeed numerical
simulations in these\nconfigurations. Accurate comparison between experim
ents and fully-resolved numerical\nsimulations will be presented in order
to bring out the relevance of the proposed algorithms.\n[1] S. Tanguy\, T.
Menard\, A. Berlemont\, A level set method for vaporizing two-phase flows
\, J.\nComput. Phys. 221 (2007) 837-853\n[2] S. Tanguy\, M. Sagan\, B. Lal
anne\, F. Couderc\, C. Colin\, Benchmarks and numerical methods\nfor the s
imulation of boiling flows. J. Comput. Phys. 264 (2014) 1-22.\n[3] L. Rued
a Villegas\, R. Alis\, M. Lepilliez\, S. Tanguy. A Level Set/Ghost Fluid M
ethod for\nboiling flows and liquid evaporation: Application to the Leiden
frost effect. J. Comp. Phys. 316\n(2016) 789-813\n[4] L. Rueda Villegas\,
S. Tanguy\, G. Castanet\, O. Caballina\, F. Lemoine. Direct Numerical Simu
lation of the impact of a droplet onto a hot surface above the Leidenfrost
temperature. Int. J. Heat Mass Transfer 104 (2017) 1090-1109\n[5] G. Hube
r\, M. Sagan\, C. Colin\, S. Tanguy. Direct Numerical Simulation of nuclea
te boiling at\nmoderate Jakob number and high microscopic contact angle. I
n preparation to be submitted in Int. J. Heat Mass Transfer\n\nhttps://ind
ico.math.cnrs.fr/event/1577/contributions/569/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/569/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Modelling atomization with phase change
DTSTART;VALUE=DATE-TIME:20161212T160000Z
DTEND;VALUE=DATE-TIME:20161212T163000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-570@indico.math.cnrs.fr
DESCRIPTION:Speakers: François-Xavier Demoulin (CORIA)\nDNS[1]\, LES [2]
and RANS [3] modelling of atomization have been developed for the last dec
ade in our\nlaboratory with a particular attention devoted on the behavior
of the interface. In particular model\nequations for the liquid-gas surfa
ce density have been proposed based on the pioneering work of Borghi\nand
Vallet [4]. The purpose of this approach is to determine the surface densi
ty that we believe is the first\norder parameter to determine the mass tra
nsfer rate\, a key future of fuel injection system. In addition to\nwell-d
eveloped procedures usually used to evaluate the vaporization rate for dis
persed spray based on\nthe resolution of Boltzmann-Williams kinetic equati
on\, our focus has been to determine the phase change\nrate for any kind o
f interface geometry not only spherical droplet. To do so the interface ca
pturing DNS\ncode Archer has been extended to handle heat and mass transfe
r at the interface based on the method\nproposed by Tanguy et al. [5]\, [6
]. From this work the turbulent mixing of a scalar quantity issued from an
\ninterface such has the vapor concentration has been studied showing the
importance of interface\nboundary layer zone on the global statistic of th
e scalar field [7]. Further works concern the extension of\ninterface capt
uring method generally based on incompressible scheme to fully compressibl
e code to\nhandle other phenomena occurring during the injection process s
uch as cavitation.\n[1] T. Menard\, S. Tanguy\, et A. Berlemont\, « Coupl
ing level set/VOF/ghost fluid methods: Validation and\napplication to 3D s
imulation of the primary break-up of a liquid jet »\, Int. J. Multiph. Fl
ow\, vol. 33\, n o\n5\, p. 510‐524\, 2007.\n[2] J. Chesnel\, J. Reveillo
n\, T. Menard\, et F. X. Demoulin\, « Large eddy simulation of liquid jet
\natomization »\, At. Sprays\, vol. 21\, n o 9\, p. 711‐736\, 2011.\n[3
] R. Lebas\, T. Menard\, P. A. Beau\, A. Berlemont\, et F. X. Demoulin\,
« Numerical simulation of primary\nbreak-up and atomization: DNS and mode
lling study »\, Int. J. Multiph. Flow\, vol. 35\, n o 3\, p. 247‐260\,\
n2009.\n[4] A. Vallet et R. Borghi\, « Modélisation Eulerienne de L’at
omisation d’un Jet Liquide »\, C R Acad Sci\nParis Sér II B\, vol. 327
\, p. 1015–1020\, 1999.\n[5] S. Tanguy\, T. Ménard\, et A. Berlemont\,
« A Level Set Method for vaporizing two-phase flows »\, J.\nComput. Phys
.\, vol. 221\, n o 2\, p. 837‐853\, 2007.\n[6] B. Duret\, T. Menard\, J.
Reveillon\, et F. X. Demoulin\, « Two phase flows DNS of evaporating liq
uid-gas\ninterface including interface regression\, using Level Set and Co
upled Level Set/VOF method »\,\nprésenté à 8th International Conferenc
e on Multiphase Flow (ICMF 2013\, 2013.\n[7] B. Duret\, G. Luret\, J. Reve
illon\, T. Menard\, A. Berlemont\, et F. X. Demoulin\, « DNS analysis of\
nturbulent mixing in two-phase flows »\, Int. J. Multiph. Flow\, vol. 40\
, n o 0\, p. 93‐105\, 2012.\n\nhttps://indico.math.cnrs.fr/event/1577/co
ntributions/570/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/570/
END:VEVENT
BEGIN:VEVENT
SUMMARY:High fidelity anisotropic adaptive FEM towards physical couplings
occurring in turbulent boiling
DTSTART;VALUE=DATE-TIME:20161212T153000Z
DTEND;VALUE=DATE-TIME:20161212T160000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-571@indico.math.cnrs.fr
DESCRIPTION:Speakers: Elie Hachem ()\nWe propose in this work an adaptive
variational multiscale method for complex multiphase\nflows with surface t
ension: applications to 3D bubble dynamics\, turbulent boiling and solid\n
quenching with experimental comparisons will be presented. A new conservat
ive level-set\nmethod is used to provide a precise position of the interfa
ces. An implicit implementation of\nthe surface tension in the context of
the Continuum Surface Force is proposed to circumvent\nthe capillary time
step restriction. The obtained system is then solved using a unified\ncomp
ressible-incompressible variational multiscale stabilized finite element m
ethod\ndesigned to handle the abrupt changes at the interface and large de
nsity and viscosity ratios.\nCombined with an a posteriori error estimator
\, we show that anisotropic mesh adaptation\nyields an accurate 3D modelin
g framework for turbulent multiphase flows with phase\nchange.\n\nhttps://
indico.math.cnrs.fr/event/1577/contributions/571/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/571/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Numerical simulation of flows with sharp interfaces by the Volume-
Of-Fluid method
DTSTART;VALUE=DATE-TIME:20161212T131500Z
DTEND;VALUE=DATE-TIME:20161212T140000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-577@indico.math.cnrs.fr
DESCRIPTION:Speakers: Stéphane Zaleski (Université Pierre et Marie Curie
(Paris))\nWe discuss recent developments in the Volume-Of-Fluid (VOF) met
hods\, such as the height\nfunction method for the approximation of the ge
ometry of the interface\, the balanced-force\nsurface tension method\, and
the methods that conserve mass and momentum at machine accuracy. Applicat
ions at high Reynolds number\,such as high speed liquid-gas flows\, and lo
w Reynolds and low Capillary numbers\, are discussed.\nProblems of enginee
ring and physical interest\, such as jet atomisation or flow in porous med
ia are investigated with these methods as will be shown.\n\nhttps://indico
.math.cnrs.fr/event/1577/contributions/577/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/577/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Locally conservative approximation of (conservative) systems writt
en in non conservation form: application to Lagrangian hydrodynamics and m
ultifluid problems
DTSTART;VALUE=DATE-TIME:20161212T163000Z
DTEND;VALUE=DATE-TIME:20161212T170000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-572@indico.math.cnrs.fr
DESCRIPTION:Speakers: Rémi Abgrall (Université de Zürich)\nSince the ce
lebrated Lax Wendroff theorem\, it is known that the right way of discreti
sing systems of hyperbolic\nequations written in conservation form is to u
se a flux formulation. However\, in many occasions\, the relevant formu-\n
lation\, from an engineering point of view\, is not to consider this conse
rvative formulation but one non conservative\nform. For example\, with sta
ndard notations\, a one fluid model writes\n[équation : voir résumé PDF
] (1)\nbut the interesting quantities are the mass\, velocity and pressure
\, which evolution is described by:\n[équation : voir résumé PDF] (2)\n
Unfortunately this form is not suitable to approximation. In the case of a
multi-fluid system\, the same problem occurs.\nIn this talk\, we will des
cribe a method to overcome this issue. It does not use any flux formulatio
n per se\, but\ncan be shown to provide the right solutions. In order to i
llustrate the method\, we will consider several examples in\nEulerian and
Lagrangian hydrodynamics\nWe will first start from the Residual Distributi
on (RD) (re-)interpretation of the Dobrev et al. scheme [1] for\nthe numer
ical solution of the Euler equations in Lagrangian form. The first ingredi
ent of the original scheme is\nthe staggered grid formulation which uses c
ontinuous node-based finite element approximations for the kinematic\nvari
ables and cell-centered discontinuous finite elements for the thermodynami
c parameters. The second ingredient\nof the Dobrev et al. scheme is an art
ificial viscosity technique applied in order to make possible the computat
ion of\nstrong discontinuities. Using a reformulation in term of RD scheme
\, we can show that the scheme is indeed locally\nconservative while the f
ormulation is stricto sensu non conservative. Using this\, we can generali
se the construction\nand develop locally conservative artificial viscosity
free schemes. To demonstrate the robustness of the proposed RD\nscheme\,
we solve several one-dimensional shock tube problems from rather mild to v
ery strong ones: we go from the\nclassical Sod problem\, to TNT explosions
(with JWL EOS) via the Collela-Woodward blast wave problem.\nIn a second
part\, we show how to extend this method to the Eulerian framework and giv
e applications on single\nfluid and multiphase problems via the five equat
ion model.\nReferences\n[1] V. Dobrev\, T. Kolev\, and R. Rieben. High ord
er curvilinear finite element methods for Lagrangian hydrodynamics.\nSIAM
J. Sci. Comput\, 34:B606–B641\, 2012.\n\nhttps://indico.math.cnrs.fr/eve
nt/1577/contributions/572/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/572/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Numerical simulation of a gas bubble collapse using the SPH-ALE me
thod
DTSTART;VALUE=DATE-TIME:20161213T140000Z
DTEND;VALUE=DATE-TIME:20161213T143000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-573@indico.math.cnrs.fr
DESCRIPTION:Speakers: Saira Pineda ()\nA multiphase model developed in SPH
-ALE is used to simulate the collapse of a gas bubble in water. This model
does not\ndiffuse the interface and guarantees the continuity of normal v
elocity and pressure at the interface between both fluids.\nThis scheme is
able to deal with interfaces of simple contact where normal velocity is c
ontinuous.\nThe model solves the mass\, momentum and energy conservation e
quations of Euler system using a non-isentropic equation\nof state for eac
h phase\, the Stiffened Gas EOS for water and the ideal gas EOS for the ga
s bubble. Both phases are\ncompressible and the phase change is not modele
d.\nA multiphase shock tube is presented for validation purpose\, with sat
isfactory results in comparison with reference\nsolutions. The dynamics of
the Rayleigh collapse of a bubble in a free-field and near a planar rigid
wall are analyzed.\nCollapse behavior\, interfacial velocities and surfac
e pressure as a function of time are analyzed for the free-field collapse\
ncase\, and in addition\, as a function of the initial bubble stand-off di
stance from the wall for the case of the bubble collapse\nnear the wall.\n
For the case of the bubble collapse near a wall\, a re-entrant jet directe
d towards the surface is observed due to the non-\nsymmetry initial config
uration. The potential damage to the surface wall is estimated by measurin
g the wall pressure.\n\nhttps://indico.math.cnrs.fr/event/1577/contributio
ns/573/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/573/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Benchmarking rotating flow with free surface deformation
DTSTART;VALUE=DATE-TIME:20161213T133000Z
DTEND;VALUE=DATE-TIME:20161213T140000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-574@indico.math.cnrs.fr
DESCRIPTION:Speakers: Laurent Martin Witkowski ()\nThe free surface deform
ation generated by a disk rotating at the bottom of a container partially
filled with fluid\nis an exciting challenge for numerical simulations. The
shape of the free surface has shown surprising\npatterns in experiments p
erformed by different research groups. However\, for many regimes (non\nax
isymmetric\, dewetted disk\, sloshing)\, an accurate comparison with numer
ical simulations is clearly\nmissing. We will present the different existi
ng regimes of such flow and show results of comparison between\ndifferent
numerical codes on few selected regimes. Some preliminary measurements on
a recent\nexperimental set up will also be presented and we will discuss t
he relevance of a benchmark on such flow.\n\nhttps://indico.math.cnrs.fr/e
vent/1577/contributions/574/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/574/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Direct Numerical Simulation of Bubbles with Adaptive Mesh Refineme
nt with Distributed Algorithms
DTSTART;VALUE=DATE-TIME:20161212T084500Z
DTEND;VALUE=DATE-TIME:20161212T093000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-575@indico.math.cnrs.fr
DESCRIPTION:Speakers: Arthur Talpaert (CEA and Ecole Polytechnique (Paris-
Saclay))\nThis talk presents the implementation of the simulation of two-p
hase flows in conditions of\nwater-cooled nuclear reactors\, at the scale
of individual bubbles. To achieve that\, we study several models for Therm
al-Hydraulic flows and we focus on a technique for the capture of the thin
interface between liquid and vapour phases. We thus review some possible
techniques for Adaptive Mesh Refinement (AMR) and provide algorithmic and
computational tools adapted to patch-based AMR\, which aim is to locally i
mprove the precision in regions of interest. More precisely\, we introduce
a patch-covering algorithm designed with balanced parallel computing in m
ind. This approach lets us finely capture changes located at the interface
\, as we show for advection test cases as well as for models with hyperbol
ic-elliptic coupling. The computations we present also include the simulat
ion of the incompressible Navier-Stokes system\, which models the shape ch
anges of the interface between two non-miscible fluids. We highlight two c
anonical test cases: the (one-phase) lid-driven cavity as well as the Rayl
eigh-Taylor instability.\n\nhttps://indico.math.cnrs.fr/event/1577/contrib
utions/575/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/575/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lagrange-Euler Lattice-Boltzmann Method And Its Application to Two
-Fluid Flows Dynamics With Possibly High Density Ratio
DTSTART;VALUE=DATE-TIME:20161213T104500Z
DTEND;VALUE=DATE-TIME:20161213T113000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-576@indico.math.cnrs.fr
DESCRIPTION:Speakers: Marie Béchereau (Ecole normale supérieure de Cacha
n (Paris-Saclay))\nTwo-fluid extensions of Lattice Boltzmann methods with
free boundaries usually consider "microscopic''\npseudopotential interface
models. In this paper\, we rather propose an interface-capturing Lattice\
nBoltzmann approach where the mass fraction variable is considered as an u
nknown and is advected.\nSeveral works have reported the difficulties of L
BM methods to deal with such two-fluid systems\nespecially for high-densit
y ratio configurations. This is due to the mixing nature of LBM\, as with
Flux\nvector splitting approaches for Finite Volume methods. We here give
another explanation of the lack of\nnumerical diffusion of Lattice Boltzma
nn approaches to accurately capture contact discontinuities. To fix\nthe p
roblem\, we propose an arbitrary Lagrangian-Eulerian (ALE) formulation of
Lattice-Boltzmann\nmethods. In the Lagrangian limit\, it allows for a prop
er separated treatment of pressure waves and\nadvection phenomenon. After
the ALE solution\, a remapping (advection) procedure is necessary to\nproj
ect the variables onto the Eulerian Lattice-Boltzmann grid.\nWe explain ho
w to derive this remapping procedure in order to get second-order accuracy
and achieve\nsharp stable oscillation-free interfaces. It has been shown
that mass fractions variables satisfy a local\ndiscrete maximum principle
and thus stay in the range $[0\,1]$. The theory is supported by numerical\
ncomputations of the free fall of an initial square block\nof a dense flui
d surrounded by a lighter fluid into a box. Figures 1 and 2 are showing th
e mass fraction\nfield of the light fluid at two successive instants. The
density ratio equal to 4 and the computational lattice\ngrid is 400x400. O
ne can appreciate the thickness of the numerical diffuse interface\, the c
apture of\ncomplex structures and the capability to compute strong changes
of free boundary topology.\nEven if our methods are currently used for in
viscid flows (Euler equations) by projecting the discrete\ndistributions o
nto equilibrium ones at each time step\, we believe that it is possible to
extend the\nframework formulation for multifluid viscous problems. This w
ill be at the aim of a next work.\n\nhttps://indico.math.cnrs.fr/event/157
7/contributions/576/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/576/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Modelling liquid-vapor phase change with metastable states
DTSTART;VALUE=DATE-TIME:20161213T080000Z
DTEND;VALUE=DATE-TIME:20161213T084500Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-578@indico.math.cnrs.fr
DESCRIPTION:Speakers: Hélène Mathis (Université de Nantes)\nWe propose
a model of liquid-vapor phase transition including metastable\nstates of t
he van der Waals Equation of State. The first part of the talk concerns\nt
he thermodynamics model. Following the second principle\, the problem boil
s\ndown to a minimization problem with constraints of the mixture energy.
This\n”static” description allows to recover the classical equilibria:
pure liquid/vapor\nstates and a coexistence state (given by the Maxwell e
qual area rule). Then\,\nwhen assuming a dependency with respect to time\,
we define a dynamical sys-\ntem with long time equilibria which are eithe
r the classical equilibria or the\nmetastable states. In a second part of
the talk\, we use the dynamical system\nas a source term of a two-phase is
othermal model. The homogeneous model\nis hyperbolic under condition. Howe
ver for smooth solutions\, we manage to\nprove that the regions of hyperbo
lically are invariant domains. We finish with\nsome numerical experiments\
, obtained by a finite volume scheme and a splitting\ntechnique to handle
the source term.\n\nhttps://indico.math.cnrs.fr/event/1577/contributions/5
78/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/578/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Numerical simulation of Faraday wave patterns
DTSTART;VALUE=DATE-TIME:20161213T100000Z
DTEND;VALUE=DATE-TIME:20161213T104500Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-579@indico.math.cnrs.fr
DESCRIPTION:Speakers: Laurette Tuckerman (Ecole Supérieure de Physique et
de Chimie Industrielle (Paris))\nIn 1831\, Faraday described the standing
wave patterns that form on\nthe surface of a layer of fluid subjected to
periodic vertical vibration.\nThese waves usually take the form of stripes
\, squares\, or hexagons.\nHowever\, other phenomena have been observed nu
merically\, such as\nquasipatterns\, supersquares\, heteroclinic cycles\,
and oscillons.\n\nUntil recently\, numerical simulation of Faraday waves w
as out of\nreach. Since 2009\, however\, we have simulated not only simpl
e wave\npatterns but also patterns which involve large-scale modulation. T
o\ndo so\, we have developed a massively parallel multiphase code\, BLUE\,
\nwhose treatment of the free surface uses a hybrid\nFront-Tracking/Level-
Set technique\, defining the interface both by a\ndiscontinuous density fi
eld on the Eulerian grid and by triangles on\nthe Lagrangian interface mes
h.\n\nWe will discuss the various Faraday wave configurations we have\nstu
died: regular square and hexagonal lattices\, patterns composed\nof spheri
cal harmonics on a vibrated drop\, and supersquares\nconsisting of a four-
by-four array of smaller squares.\n\nhttps://indico.math.cnrs.fr/event/157
7/contributions/579/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/579/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Diffuse interfaces with compressible fluids\, phase transition and
capillarity
DTSTART;VALUE=DATE-TIME:20161213T084500Z
DTEND;VALUE=DATE-TIME:20161213T093000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-580@indico.math.cnrs.fr
DESCRIPTION:Speakers: Richard Saurel (Université Aix-Marseille)\nConventi
onal models of capillary fluids with phase transition consider linked the
rmodynamics and capillarity. Such coupling has serious consequences\, such
as:\n\n- sound propagation\, undefined in some critical regions\,\n\n- ve
ry thin interfaces\, causing serious issues in practical computations.\n\n
In the present talk an approached based on hyperbolic systems with relaxat
ion is promoted to solve interfaces with phase transition and surface tens
ion. The method deals with arbitrary pressure and density jumps. \n\nThe d
iffuse interface model consists in a set of balance equations of mass for
each phase and momentum and energy for the mixture. When simple contact is
considered (in the absence of heat diffusion)\, a volume fraction equatio
n is needed as well. In this frame each phase is compressible and governed
by its own (convex) equation of state\, preserving sound propagation. The
two equations of state are rendered compatible through appropriate consta
nts determined from the phase diagram. Phase change is modeled through Gib
bs free energy relaxation terms. Capillarity is modelled through mass frac
tion gradients at interfaces and is consequently decoupled of thermodynami
cs.\nExamples of cavitating\, flashing and boiling flows with and without
shocks are shown. \n\nLe Métayer\, O.\, & Saurel\, R. (2016). The Noble-A
bel Stiffened-Gas equation of state. Physics of Fluids 28(4)\, 046102\n\n
Saurel\, R.\, Le Metayer\, O. and Boivin\, P. (2016) A general formulation
for cavitating\, boiling and evaporating flows. Computers and Fluids 128\
, 53-64\n\nLe Martelot\, S.\, Saurel\, R. and Nkonga\, B. (2014) Toward th
e direct numerical simulation of boiling flows. Int. J. of Multiphase
Flows 77\, 62-78\n\nLe Martelot\, S.\, Nkonga\,B. and Saurel\, R. (2013) L
iquid and liquid–gas flows at all speeds\, Journal of Computational Phys
ics 255(15)\, 53-82\n\nPetitpas\, F.\, Massoni\, J.\, Saurel\, R.\, Lapebi
e\, E. and Munier\, L. (2009) Diffuse interface model for high speed cavit
ating underwater systems. Int. J. Multiphase Flow 35\, 747-759\n\nSaurel\,
R.\, Petitpas\, F. and Abgrall\, R. (2008) Modeling phase transition in m
etastable liquids. Application to flashing and cavitating flows. Journal o
f Fluid Mechanics\, 607:313-350\n\nhttps://indico.math.cnrs.fr/event/1577/
contributions/580/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/580/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Effervescence in champagne and sparkling wines: Recent advances an
d future prospects
DTSTART;VALUE=DATE-TIME:20161212T080000Z
DTEND;VALUE=DATE-TIME:20161212T084500Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-581@indico.math.cnrs.fr
DESCRIPTION:Speakers: Gérard Liger-Belair (CNRS)\nBubbles in a glass of c
hampagne may seem like the acme of frivolity to most of people\, but in fa
ct they may rather be considered as a fantastic playground for any fluid p
hysicist. In a glass of champagne\, about a million bubbles will nucleate
and rise if you resist drinking from your flute. The so-called effervescen
ce process\, which enlivens champagne and sparkling wines tasting\, is the
result of the complex interplay between carbon dioxide (CO2) dissolved in
the liquid phase\, tiny air pockets trapped within microscopic particles
during the pouring process\, and some both glass and liquid properties. Th
e journey of yeast-fermented CO2 is reviewed (from its progressive dissolu
tion in the liquid phase during the fermentation process\, to its progress
ive release in the headspace above glasses). The physicochemical processes
behind the nucleation\, and rise of gaseous CO2 bubbles\, under standard
tasting conditions\, have been gathered hereafter. Moreover\, when a bubbl
e reaches the air-champagne interface\, it ruptures\, projecting a multitu
de of tiny droplets in the air. Based on the model experiment of a single
bubble bursting in simple liquids\, we depict each step of this process\,
from bubble bursting to droplet evaporation. In particular\, we demonstrat
e how damping action of viscosity produces faster and smaller droplets and
more generally how liquid properties enable to control the bubble burstin
g aerosol characteristics. We demonstrate that compared to a still wine\,
champagne fizz drastically enhances the transfer of liquid into the atmosp
here. Conditions on bubble radius and wine viscosity that optimize aerosol
evaporation are provided. These results pave the way towards the fine tun
ing of aerosol characteristics and flavor release during sparkling wine ta
sting\, a major issue of the sparkling wine industry.\n\nhttps://indico.ma
th.cnrs.fr/event/1577/contributions/581/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/581/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Applications of the Front-Tracking algorithm of TrioCFD to turbule
nt bubbly flows in plane channels
DTSTART;VALUE=DATE-TIME:20161212T140000Z
DTEND;VALUE=DATE-TIME:20161212T144500Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-582@indico.math.cnrs.fr
DESCRIPTION:Speakers: Guillaume Bois (CEA (Paris-Saclay))\nThe Front-Track
ing method has been implemented in TrioCFD and improved over the last\ndec
ade. It has been widely used on large parallel architectures to study inco
mpressible two-\nphase flows. The permanent increase in computing capabili
ties allows to perform simulations\nof fully turbulent bubbly flows in rel
atively small periodic domains. This talk will be organized\nin two parts.
The numerical method used to perform Front-Tracking simulations will be\n
presented. The code is capable to deal with phase change and specific Ghos
t-Fluid Methods\nhave been implemented to guarantee a great accuracy of th
e solution\, even in the presence\nof large jumps and phase change.\nThen\
, recent calculations on adiabatic two-phase turbulent bubbly flows will b
e presented.\nAveraged results are analyzed in great details in order to b
etter understand the dominant\nprocesses in the exchange mechanisms at the
interface and in the modulation of turbulence\nby the vapor inclusions an
d their wakes. Preliminary results and suggestions for the two-fluid\nmode
l will conclude the presentation.\nThis work was granted access to the HPC
resources of TGCC under the allocation 20XX-\nt20142b7239 made by GENCI.\
n\nhttps://indico.math.cnrs.fr/event/1577/contributions/582/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/582/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Simulation of free surface fluids in incompressible dynamique
DTSTART;VALUE=DATE-TIME:20161212T103000Z
DTEND;VALUE=DATE-TIME:20161212T110000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-583@indico.math.cnrs.fr
DESCRIPTION:Speakers: Dena Kazerani (Laboratoire Jacques-Louis Lions (UPMC
\, Paris 6))\nIn this work\, we present a numerical scheme for solving fre
e surface flows. The free surface is modeled using the level-set formulati
on. Besides\, the mesh is anisotropic and adapted at each iteration. This
adaptation allows us to obtain a precise approximation for the free- surfa
ce location. In addition\, it enables us to solve the time-discretized flu
id equation only on the fluid domain. The fluid here is considered incompr
essible. Therefore\, its motion is described by the incompressible Navier
–Stokes equation which is temporally discretized using the method of cha
racteristics and is solved at each time iteration by a first order Lagrang
e–Galerkin method. The level-set function representing the free surface
satisfies an advection equation which is also solved using the method of c
haracteristics. The algorithm is completed by some intermediate steps like
the construction of a convenient initial level-set function (redistancing
) as well as the construction of a convenient flow for the level-set advec
tion equation. Finally\, some numerical results are presented.\n\nhttps://
indico.math.cnrs.fr/event/1577/contributions/583/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/583/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Simulations of a heated fluid at low Mach number: modelling of pha
se transition and numerical strategies
DTSTART;VALUE=DATE-TIME:20161213T143000Z
DTEND;VALUE=DATE-TIME:20161213T150000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-584@indico.math.cnrs.fr
DESCRIPTION:Speakers: Bérénice Grec ()\nThermohydraulic codes used in in
dustry are based on the resolution of compressible Navier-Stokes equations
in which acoustic waves are taken into account. This allows to describe f
luid flows at any Mach number. However\, many difficulties may arise in te
rms of CPU time\, robustness and accuracy in the low Mach number regime.\n
In this regime\, an asymptotic expansion with respect to the Mach number l
eads to simpler models.\nThus\, the strategy of our work is to derive\, in
vestigate and simulate a system of PDE taking into account phase transitio
n in the low Mach number regime but with possible high heat transfers.\nMo
re precisely\, we focus on the choice of the equation of state and its par
ameters\, with emphasis on the gain due to the low Mach number hypothesis\
, \nand we present preliminary 2D numerical simulations with FreeFem++ sho
wing the robustness of the approach.\nThis is a joint work with S. Dellach
erie\, G. Faccanoni and Y. Penel.\n\nhttps://indico.math.cnrs.fr/event/157
7/contributions/584/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/584/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Numerical simulations of gas/vapor bubble oscillations
DTSTART;VALUE=DATE-TIME:20161212T100000Z
DTEND;VALUE=DATE-TIME:20161212T103000Z
DTSTAMP;VALUE=DATE-TIME:20180722T083136Z
UID:indico-contribution-1577-585@indico.math.cnrs.fr
DESCRIPTION:Speakers: Daniel FUSTER (Institut D'Alembert\, UPMC-CNRS 75005
Paris)\nIn this work we numerically investigate the effect of heat and ma
ss transfer on the dynamic response of gas-vapor bubbles. The numerical s
olution of the full non-linear 1D equations is compared with the analytica
l solution of the equations obtained for the oscillation of an spherical g
as/vapor bubble in\nresponse of a weak pressure perturbation (linear solut
ion). For a system with known gas/vapor/liquid properties\, we identify v
arious oscillation regimes as a function of an nondimensional oscillation
frequency (e.g. the bubble's Peclet number) and the vapor content. Even at
very low frequencies\, there exist regimes where transient diffusion effe
cts arise due to heat diffusion in the surrounding liquid and also due to
vapor mass diffusion inside the bubble. These phenomena restrict the appli
cability of the commonly-adopted assumption of full-equilibrium conditions
inside the bubble. Simulations of the oscillation of bubbles for strong
perturbations shows that non-linear effects restrict even further the rang
e of applicability of the isothermal equilibrium model when the vapor cont
ent becomes larger than a critical value.\n\nhttps://indico.math.cnrs.fr/e
vent/1577/contributions/585/
LOCATION:Institut Henri Poincaré Amphi Hermite
URL:https://indico.math.cnrs.fr/event/1577/contributions/585/
END:VEVENT
END:VCALENDAR