Abstract
The modelling of diffusive terms in particle methods is a delicate matter and several models were proposed in the literature to take such terms into account. The diffusion velocity method (DVM), originally designed for the diffusion of passive scalars, turns diffusive terms into convective ones by expressing them as a divergence involving a so-called diffusion velocity. In this paper, DVM is extended to the diffusion of vectorial quantities in the three-dimensional Navier–Stokes equations, in their incompressible, velocity–vorticity formulation. The integration of a large eddy simulation (LES) turbulence model is investigated and a DVM general formulation is proposed. Either with or without LES, a novel expression of the diffusion velocity is derived, which makes it easier to approximate and which highlights the analogy with the original formulation for scalar transport. From this statement, DVM is then analysed in one dimension, both analytically and numerically on test cases to point out its good behaviour.
Similar content being viewed by others
References
Beale JT, Majda A (1985) High order accurate vortex methods with explicit velocity kernels. J Comput Phys 58(2): 188–208, doi:10.1016/0021-9991(85)90176-7, http://www.sciencedirect.com/science/article/B6WHY-4DDR2KW-47/2/a1fa01ee4805464c4e45e0a3dededa59
Beaudoin A, Huberson S, Rivoalen E (2003) Simulation of anisotropic diffusion by means of a diffusion velocity method. J Comput Phys 186(1): 122–135, doi:10.1016/S0021-9991(03)00024-X, http://www.sciencedirect.com/science/article/B6WHY-47YPK0S-2/2/d4dea06c15c422a20f525b6f67683cd4
Bonet J, Lok TS (1999) Variational and momentum preservation aspects of smooth particle hydrodynamic formulations. Comput Methods Appl Mech Eng 180(1–2): 97–115, doi:10.1016/S0045-7825(99)00051-1, http://www.sciencedirect.com/science/article/pii/S0045782599000511
Chatelain P, Backaert S, Winckelmans G, Kern S (2013) Large eddy simulation of wind turbine wakes. Flow Turbul Combust 91(3):587–605. doi:10.1007/s10494-013-9474-8
Chertock A, Levy D (2001) Particle methods for dispersive equations. J Comput Phys 171(2): 708–730, doi:10.1006/jcph.2001.6803, http://www.sciencedirect.com/science/article/pii/S0021999101968032
Chertock A, Levy D (2002) A particle method for the KdV equation. J Sci Comput 17(1—-4):491–499. doi:10.1023/A:1015106210404
Chorin AJ (1973) Numerical study of slightly viscous flow. J Fluid Mech 57(4):785–796. doi:10.1017/S0022112073002016
Christiansen J (1997) Numerical simulation of hydrodynamics by the method of point vortices. J Comput Phys 135(2):189–197, doi:10.1006/jcph.1997.5701, http://www.sciencedirect.com/science/article/pii/S0021999197957016
Colagrossi A, Landrini M (2003) Numerical simulation of interfacial flows by smoothed particle hydrodynamics. J Comput Phys 191(2): 448–475, doi:10.1016/S0021-9991(03)00324-3, http://www.sciencedirect.com/science/article/pii/S0021999103003243
Cortez R (1997) Convergence of high-order deterministic particle methods for the convection-diffusion equation. Commun Pure Appl Math 50(L):1235–1260
Cottet GH, Poncet P (2004) Advances in direct numerical simulations of 3d wall-bounded flows by vortex-in-cell methods. J Comput Phys 193(1):136–158, doi:10.1016/j.jcp.2003.08.025, http://www.sciencedirect.com/science/article/pii/S0021999103004248
Degond P, Mas-Gallic S (1989) The weighted particle method for convection-diffusion equations. Part I: The case of an isotropic viscosity. Math Comput 53(188):485–507. doi:10.2307/2008716
Degond P, Mustieles FJ (1990) A deterministic approximation of diffusion equations using particles. SIAM J Sci Stat Comput 11(2): 293–310, doi:10.1137/0911018, http://link.aip.org/link/?SCE/11/293/1
Dynnikov Y, Dynnikova G (2011) Numerical stability and numerical viscosity in certain meshless vortex methods as applied to the navier-stokes and heat equations. Comput Math Math Phys 51:1792–1804. doi:10.1134/S096554251110006X
Dynnikova G (2011) Calculation of three-dimensional flows of an incompressible fluid based on a dipole representation of vorticity. Dokl Phys 56(3):163–166. doi:10.1134/S1028335811030025
Eldredge JD, Leonard A, Colonius T (2002) A general deterministic treatment of derivatives in particle methods. J Comput Phys 180(2): 686–709. doi:10.1006/jcph.2002.7112, http://www.sciencedirect.com/science/article/B6WHY-46G47HY-F/2/a01a8d80f16f9d8f1275850e03158789
Fronteau J, Combis P (1984) A lie admissible method of integration of folkler-plank equations with non linear coefficients (exact and numerical solutions). Hadronic J 7:911–930
Gambino G, Lombardo M, Sammartino M (2009) A velocity-diffusion method for a lotka-volterra system with nonlinear cross and self-diffusion. Appl Numer Math 59(5):1059–1074. doi:10.1016/j.apnum.2008.05.002, http://www.sciencedirect.com/science/article/pii/S0168927408000925
Grant J, Marshall J (2005) Diffusion velocity for a three-dimensional vorticity field. Theor Comput Fluid Dyn 19:377–390. doi:10.1007/s00162-005-0004-8
Guvernyuk S, Dynnikova G (2007) Modeling the flow past an oscillating airfoil by the method of viscous vortex domains. Fluid Dyn 42:1–11. doi:10.1134/S0015462807010012
Kempka S, Strickland J (1993) A method to simulate viscous diffusion of vorticity by convective transport of vortices at a non-solenoidal velocity. Technical Report SAND-93-1763, Sandia Laboratory, http://www.osti.gov/energycitations/servlets/purl/10190654-dLAMwy/native/
Lacombe G (1999) Analyse d’une équation de vitesse de diffusion. Comptes Rendus de l’Académie des Sciences—Series I—Mathematics, vol 329, no 5, pp 383–386. doi:10.1016/S0764-4442(00)88610-3, URL http://www.sciencedirect.com/science/article/pii/S0764444200886103
Lacombe G, Mas-Gallic S (1999) Presentation and analysis of a diffusion-velocity method. ESAIM Proc 7:225–233. doi:10.1051/proc:1999021
Leonard A (1980) Vortex methods for flow simulation. J Comput Phys 37(3): 289–335. doi:10.1016/0021-9991(80)90040-6, http://www.sciencedirect.com/science/article/B6WHY-4DD1RFV-1T/2/68d5c3a0ce2f03f137353c200b9df3e8
Lions PL, Mas-Gallic S (2001) Une méthode particulaire déterministe pour des équations diffusives non linéaires. Comptes Rendus de l’Académie des Sciences—Series I—Mathematics 332(4), pp 369–376. doi:10.1016/S0764-4442(00)01795-X, http://www.sciencedirect.com/science/article/B6VJ2-42JYXCG-J/2/7ba004247cb104c46218511bd49d8956
Liu WK, Jun S, Zhang YF (1995) Reproducing kernel particle methods. Int J Numer Methods Fluids 20(8—-9):1081–1106. doi:10.1002/fld.1650200824
Liu WK, Chen Y, Uras R, Chang CT (1996) Generalized multiple scale reproducing kernel particle methods. Comput Methods Appl Mech Eng 139(1–4): 91–157. doi:10.1016/S0045-7825(96)01081-X, http://www.sciencedirect.com/science/article/pii/S004578259601081X
Mansfield JR (1997) A dynamic Lagrangian Large Eddy Simulation scheme for the vorticity transport equation. PhD thesis, John Hopkins University, Baltimore, Maryland, USA
Mansfield JR, Knio OM, Meneveau C (1996) Towards lagrangian large vortex simulation. ESAIM Proc 1:49–64. doi:10.1051/proc:1996019
Mansfield JR, Knio OM, Meneveau C (1998) A dynamic LES scheme for the vorticity transport equation: Formulation and a priori tests. J Comput Phys 145(2): 693–730. doi:10.1006/jcph.1998.6051, http://www.sciencedirect.com/science/article/B6WHY-45J58W3-2N/2/ea859704c5e2194b3645e0ad3bcbff90
Mas-Gallic S (1999) A presentation of the diffusion velocity method. In: Leach P, Bouquet S, Rouet JL, Fijalkow E (eds) Dynamical Systems, Plasmas and Gravitation, Lecture Notes in Physics. Springer, Berlin, pp 74–81. doi:10.1007/BFb0105914
Meneveau C, Katz J (2000) Scale-invariance and turbulence models for large-eddy simulation. Ann Rev Fluid Mech 32(1): 1–32. doi:10.1146/annurev.fluid.32.1.1, http://www.annualreviews.org/doi/abs/10.1146/annurev.fluid.32.1.1, http://www.annualreviews.org/doi/pdf/10.1146/annurev.fluid.32.1.1
Milane R, Nourazar S (1995) On the turbulent diffusion velocity in mixing layer simulated using the vortex method and the subgrid scale vorticity model. Mech Res Commun 22(4): 327–333. doi:10.1016/0093-6413(95)00032-M, http://www.sciencedirect.com/science/article/pii/009364139500032M
Milane R, Nourazar S (1997) Large-eddy simulation of mixing layer using vortex method: Effect of subgrid-scale models on early development. Mech Res Commun 24(2): 215–221. doi:10.1016/S0093-6413(97)00015-3, http://www.sciencedirect.com/science/article/pii/S0093641397000153
Milane RE (2004) Large eddy simulation (2d) using diffusion-velocity method and vortex-in-cell. Int J Numer Methods Fluids 44(8): 837–860. doi:10.1002/fld.673
Mustieles FJ (1990) L’équation de Boltzmann des semiconducteurs. Étude mathématique et simulation numérique. PhD thesis, école Polytechnique
Mycek P, Pinon G, Germain G, Rivoalen E (2013) A self-regularising DVM-PSE method for the modelling of diffusion in particle methods. Comptes Rendus Mécanique 341(9–10):709–714. doi:10.1016/j.crme.2013.08.002, http://www.sciencedirect.com/science/article/pii/S1631072113001137
Ogami Y (1999) Simulation of heat-vortex interaction by the diffusion velocity method. ESAIM Proc 7:313–324. doi:10.1051/proc:1999029
Ogami Y, Akamatsu T (1991) Viscous flow simulation using the discrete vortex model-the diffusion velocity method. Comput Fluids 19(3–4): 433–441. doi:10.1016/0045-7930(91)90068-S, http://www.sciencedirect.com/science/article/pii/004579309190068S
Oger G, Doring M, Alessandrini B, Ferrant P (2007) An improved sph method: Towards higher order convergence. J Comput Phys 225(2): 1472–1492. doi:10.1016/j.jcp.2007.01.039, http://www.sciencedirect.com/science/article/pii/S0021999107000630
Pinon G, Mycek P, Germain G, Rivoalen E (2012) Numerical simulation of the wake of marine current turbines with a particle method. Renew Energy 46(0): 111–126. doi:10.1016/j.renene.2012.03.037, http://www.sciencedirect.com/science/article/pii/S0960148112002418
Rivoalen E, Huberson S (1999) Numerical simulation of axisymmetric viscous flows by means of a particle method. J Comput Phys 152(1): 1–31. doi:10.1006/jcph.1999.6210, http://www.sciencedirect.com/science/article/pii/S0021999199962101
Rivoalen E, Huberson S, Hauville F (1997) Simulation numérique des équations de Navier-Stokes 3D par une méthode particulaire. Comptes Rendus de l’Académie des Sciences—Series IIB—Mechanics-Physics-Chemistry-Astronomy, vol 324, no 9, pp. 543–549. doi:10.1016/S1251-8069(97)83187-9, http://www.sciencedirect.com/science/article/B6VJ4-3WNN24T-J/2/059096dfb12885bea81e404865dd6a63
Sagaut P (2006) Large Eddy Simulation for Incompressible Flows: an Introduction. Scientific Computation, Springer. http://books.google.fr/books?id=ODYiH6RNyoQC
Schrader B, Reboux S, Sbalzarini IF (2010) Discretization correction of general integral pse operators for particle methods. J Comput Phys 229(11): 4159–4182. doi:10.1016/j.jcp.2010.02.004, http://www.sciencedirect.com/science/article/B6WHY-4YCWNPG-2/2/674fb3a5ba4f77a55861454096dafc5c
Strickland JH, Kempka SN, Wolfe WP (1996) Viscous diffusion of vorticity using the diffusion velocity concept. ESAIM Proc 1:135–151. doi:10.1051/proc:1996033
Winckelmans G, Cocle R, Dufresne L, Capart R (2005) Vortex methods and their application to trailing wake vortex simulations. Comptes Rendus Physique 6(4–5):467–486. doi:10.1016/j.crhy.2005.05.001, http://www.sciencedirect.com/science/article/pii/S1631070505000575
Acknowledgments
The authors would like to thank the Haute-Normandie Regional Council and the Institut Français de Recherche pour l’Exploitation de la Mer (IFREMER) for their financial support of co-financed PhD theses, as well as the Réseau d’Hydrodynamique Normand (RHYNO). The authors would also like to thank the Centre des Ressources Informatiques de HAute-Normandie (CRIHAN) for their available numerical computation resources.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Raphaèle Herbin.
Appendices
Appendix A: Vectorial DVM developments
The aim of this appendix is to show that \(\mathrm {B}\) can be expressed as in Eq. (16). First, from the definition (8), it follows that:
were repeated indices (here \(p\)) indicate a sum over those indices. Using the unit vector \(\mathbf {e}=\varvec{\omega }/|\varvec{\omega }|\), the previous equation becomes
One can then make the following transformations:
whose last term vanishes since \(e_p \partial e_p/\partial x_i = \frac{1}{2} \partial (e_pe_p)/\partial x_i = 0\). Finally, \(B_{ij}\) becomes:
Appendix B: Analytic expressions of the Gaussian kernel and its Fourier transform
The purpose of this appendix is to determine the explicit expressions of the one-dimensional Gaussian kernel and its Fourier transform, for any order of accuracy. The approach is mostly inspired by reproducing kernel particle method (RKPM) techniques [see for example Liu et al. (1995, 1996)]. One should remember that the following derivation is restricted to Gaussian kernels, and aims at providing expressions that are explicit and simpler than the more general recurrence relations proposed in the existing literature.
1.1 Notations
Let \(\zeta _{m, \varepsilon }\) be the interpolation kernel of order \(r=2(m+1)\), where \(\varepsilon \) is the mollifying parameter, issuing from the function \(\zeta _m\) [cf. Eq. (28)]. Since the kernel \(\zeta _{m, \varepsilon }\) (resp. its Fourier transform \({\hat{\zeta }}_{m, \varepsilon }\)) can easily be deduced from \(\zeta _m\) (resp. \({\hat{\zeta }}_{m}\)), whose expression is much simpler, the proof will be performed on \(\zeta _m\) and \({\hat{\zeta }}_{m}\). The 2nd order function \(\zeta _{0}\) is defined by:
It follows that
1.2 Hermite polynomials
Let \(H_n\) be the \(n\)th Hermite polynomial, whose degree is \(n\), \(n\ge 0\). \(H_n\) is classically defined by:
These polynomials have an explicit expression, which is here split up according to the parity of \(n\):
1.3 Hermite polynomials with parameter \(a\)
Hermite polynomials can easily be generalized to Gaussian functions with scaling parameter \(a>0\). Let \(H^*_n(x,a)\) be the \(n\)-th generalized Hermite function of parameter \(a\), defined by:
It is related to \(H_n\) by the following relation:
1.4 Hermitian linear combination
The \(r\)th order function \(\zeta _m\) is built from a Gaussian template as follows (Eldredge et al. 2002):
where the \(\gamma ^{[m]}_n\) are \(m\) related coefficients determined by the moment conditions. This means that \(\zeta _m\) is the product of \(\frac{\mathrm{e}^{-x^2}}{\sqrt{\pi }}\) with an even polynomial (i.e. of even powers only) of degree \(2m\). The proof relies on the fact that \(\zeta _m\) can be written as a linear combination of the \(m+1\) first even derivatives of \(\zeta _{0}\) (including the zero-th derivative):
As a matter of fact, \(\zeta _{0}^{(2n)}\) are even polynomials of degree \(2n\) multiplied by the Gaussian function \(\frac{\mathrm{e}^{-x^2}}{\sqrt{\pi }}\). It can be easily shown that any family of \(m+1\) even polynomials of degrees \(0\) to \(2m\) forms a basis for the \(2m\)th degree even polynomials.
By taking the Fourier transform of (67), one obtains:
1.5 Moment conditions
The moment conditions [cf., amongst others, Beale and Majda (1985), Liu et al. (1995), Eldredge et al. (2002)] are:
However, thanks to the choice of even functions (cf. Eq. (66)), odd moment conditions are naturally respected. Moment conditions (69b) thus reduce to:
In the Fourier domain, those conditions are, respectively, equivalent to:
1.6 Expression of the \(c_n\)
Using (68) evaluated at \(k=0\), it follows directly from (71a) that \(c_0=1\).
In addition, for \(1 \le p \le m\), \({\hat{\zeta }}_{m}^{(2p)}(k)\) is given by:
-
If \(n>p\), \({\hat{\phi }}_{p,n}(0) = 0\) and thus
$$\begin{aligned} {\hat{\zeta }}_{m}^{(2p)}(0) = \sum _{n=0}^{p} (-1)^n c_n {\hat{\phi }}_{p,n}(0). \end{aligned}$$(73) -
If \(n \le p\), using Leibniz rule,
$$\begin{aligned} {\hat{\phi }}_{p,n}(k) = \sum _{j=0}^{2p} \left( {\begin{array}{c}2p\\ j\end{array}}\right) \frac{\mathrm{d}^{j}}{\mathrm{d}k^{j}}\left( k^{2n} \right) \frac{\mathrm{d}^{2p-j}}{\mathrm{d}k^{2p-j}} \left( {\hat{\zeta }}_{0}(k) \right) \!, \end{aligned}$$(74)which, evaluated at \(k=0\), reduces to
$$\begin{aligned} {\hat{\phi }}_{p,n}(0) = \left( {\begin{array}{c}2p\\ 2n\end{array}}\right) (2n)! \, {\hat{\zeta }}_{0}^{\left( 2(p-n) \right) }(0) = \frac{(2p)!}{\left[ 2(p-n) \right] !} {\hat{\zeta }}_{0}^{\left( 2(p-n) \right) }(0). \end{aligned}$$(75)Thanks to Hermite polynomials, it follows that, for any \(q>0\),
$$\begin{aligned} {\hat{\zeta }}_{0}^{2q}(0) = \left( \frac{\mathrm{d}^{2q}}{\mathrm{d}k^{2q}} \mathrm{e}^{-k^2/4} \right) _{k=0} = \left( \frac{1}{2^{2q}} \mathrm{e}^{-k^2/4} H_{2n}(k) \right) _{k=0} = \frac{(-1)^q}{2^{2q}} \frac{(2q)!}{q!} \end{aligned}$$(76)and, combined with Eqs. (75) and (76), Eq. (73) becomes
$$\begin{aligned} {\hat{\zeta }}_{m}^{(2p)}(0) = \frac{(-1)^p (2p)!}{2^{2p}} \sum _{n=0}^{p} \frac{2^{2n}}{(p-n)!} c_n . \end{aligned}$$(77)
Condition (71b) implies
which eventually leads to the following recursive relation:
By a simple recursive induction, it can be shown that the sequence \((c_n)_{n \ge 0}\) has the following explicit definition:
1.7 Consequences on the kernel and its Fourier transform
Injecting (80) into (67) leads to
and, using Hermite polynomials to explicitly express \(\zeta _{0}^{(2n)}(x)\), a general analytic expression for a Gaussian kernel of any order \(r\) is obtained:
where
This explicit expression is not very helpful as such. However, it follows from (81) that \(\zeta _m\) is defined from \(\zeta _{m-1}\) and the \(2m\)-th derivative of \(\zeta _{0}\):
Using again Hermite polynomials to explicitly express \(\zeta _{0}^{(2m)}(x)\), the values of the \(\gamma ^{[m]}_n \in \mathbb {Q}\), which entirely define \(\zeta _m\), can thus be incrementally induced from the previous order kernel:
In fact, the explicit expression (82) is more interesting and useful in the Fourier domain. By injecting (80) into (68), one obtains:
and thus
Those results are very interesting in the sense that they provide explicit expressions of Gaussian kernels of any order of accuracy. In addition, their Fourier transform have an incremental explicit definition and satisfy the condition \({\hat{\zeta }}_{m}(k) \le {\hat{\zeta }}_{m}(0)\), which is the condition required by Cortez (1997) for stability.
1.8 Particle approximation
In the sequel, the particle approximation of the Gaussian patch \(c(x,t)= \exp \left( -x^2/(4\nu \,t)\right) /\sqrt{4\pi \nu \, t}\), presented in Sect. 4, is analysed.
In an effort to simplify notations, the time dependence of some functions may be omitted in the sequel, although it should be present both in the function and the Fourier domains. Let \(\sigma =\sqrt{4\nu t}\) and \(\alpha =\sqrt{\varepsilon ^2 + \sigma ^2}\). Then
and thus
Using Hermite polynomials,
with
Likewise, since \(\langle \nabla c \rangle =\mathcal {F}^{-1} \left[ ik {\hat{\zeta }}_{m, \varepsilon } \hat{c} \right] \) it follows that:
with
1.9 Consequences on the error
The analytic exact expressions of \(c\), \(\nabla c\) and \(u_d\) are given by:
and
It follows that
and
Expressing \(\mathcal {E}_{\langle c \rangle }\), \(\mathcal {E}_{\langle \nabla c \rangle }\) and \(\mathcal {E}_{\langle u_d \rangle }\) from these last three expressions is straightforward.
It should be noted that \(P_m\) and \(Q_m\) being polynomials with degree \(2m\), the error \(\mathcal {E}_{\langle u_d \rangle }\) is spatially constant for the 2nd order (\(r=2\) and thus \(m=0\)) approximation, at any given time \(t>0\).
Rights and permissions
About this article
Cite this article
Mycek, P., Pinon, G., Germain, G. et al. Formulation and analysis of a diffusion-velocity particle model for transport-dispersion equations. Comp. Appl. Math. 35, 447–473 (2016). https://doi.org/10.1007/s40314-014-0200-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40314-014-0200-5
Keywords
- Diffusion velocity method
- Particle method
- Fourier analysis
- Transport-dispersion equations
- Navier–Stokes equations