Christophe Berthon

A relaxation method for two-phase flow models with hydrodynamic closure law

Summary.This paper is devoted to the numerical approximation of the solutions of a system of conservation laws arising in the modeling of two-phase flows in pipelines. The PDEs are closed by two highly nonlinear algebraic relations, namely, a pressure law and a hydrodynamic one. The severe nonlinearities encoded in these laws make the classical approximate Riemann solvers virtually intractable at a reasonable cost of evaluation. We propose a strategy for relaxing solely these two nonlinearities. The relaxation system we introduce is of course hyperbolic but all associated eigenfields are linearly degenerate. Such a property not only makes it trivial to solve the Riemann problem but also enables us to enforce some further stability requirements, in addition to those coming from a Chapman-Enskog analysis. The new method turns out to be fairly simple and robust while achieving desirable positivity properties on the density and the mass fractions. Extensive numerical evidences are provided.

An HLLC Scheme to Solve The M1 Model of Radiative Transfer in Two Space Dimensions

The M1 radiative transfer model is considered in the present work in order to simulate the radiative fields and their interactions with the matter. The model is governed by an hyperbolic system of conservation laws supplemented by relaxation source terms. Several difficulties arise when approximating the solutions of the model; namely the positiveness of the energy, the flux limitation and and the limit diffusion behavior have to be satisfied. An HLLC scheme is exhibited and it is shown to satisfy all the required properties. A particular attention is payed concerning the approximate extreme waves. These approximations are crucial to obtain an accurate scheme. The extension to the full 2D problem is proposed. It satisfies, once again, all the expected properties. Numerical experiments are proposed. They show that the considered scheme is actually less diffusive than the currently used numerical methods.

A Positive Preserving High Order VFRoe Scheme for Shallow Water Equations: A Class of Relaxation Schemes

The VFRoe scheme has been recently introduced by Buffard, Gallouet, and Herard [Comput. Fluids, 29 (2000), pp. 813-847] to approximate the solutions of the shallow water equations. One of the main interests of this method is to be easily implemented. As a consequence, such a scheme appears as an interesting alternative to other more sophisticated schemes. The VFRoe methods perform approximate solutions in good agreement with the expected ones. However, the robustness of this numerical procedure has not been proposed. Following the ideas introduced by Jin and Xin [Comm. Pure Appl. Math., 45 (1995), pp. 235-276], a relevant relaxation method is derived. The interest of this relaxation scheme is twofold. In the first hand, the relaxation scheme is shown to coincide with the considered VFRoe scheme. In the second hand, the robustness of the relaxation scheme is established, and thus the nonnegativity of the water height obtained involving the VFRoe approach is ensured. Following the same idea, a family of relaxation schemes is exhibited. Next, robust high order slope limiter methods, known as MUSCL reconstructions, are proposed. The final scheme is obtained when considering the hydrostatic reconstruction to approximate the topography source terms. Numerical experiments are performed to attest the interest of the procedure.

Efficient well-balanced hydrostatic upwind schemes for shallow-water equations

The proposed work concerns the numerical approximations of the shallow-water equations with varying topography. The main objective is to introduce an easy and systematic technique to enforce the well-balance property and to make the scheme able to deal with dry areas. To access such an issue, the derived numerical method is obtained by involving the free surface instead of the water height and this produces the scheme well-balanced independently from the numerical flux function associated with the homogeneous problem. As a consequence, we obtain an easy well-balanced scheme which preserves non-negative water height. When compared with the well-known hydrostatic reconstruction, the presented topography discretization does not involve any max function known to introduce some numerical errors as soon as the topography admits very strong variations or discontinuities. A second-order MUSCL accurate reconstruction is adopted. The proposed hydrostatic upwind scheme is next extended for considering 2D simulations performed over unstructured meshes. Several 1D and 2D numerical experiments are performed to exhibit the relevance of the scheme.

Numerical approximations of the 10-moment Gaussian closure

We propose a numerical scheme to approximate the weak solutions of the 10-moment Gaussian closure. The moment Gaussian closure for gas dynamics is governed by a conservative hyperbolic system supplemented by entropy inequalities whose solutions satisfy positiveness of density and tensorial pressure. We consider a Suliciu-type relaxation numerical scheme to approximate the solutions. These methods are proved to satisfy all the expected positiveness properties and all the discrete entropy inequalities. The scheme is illustrated by several numerical experiments.

Robustness of MUSCL schemes for 2D unstructured meshes

We consider second-order accuracy MUSCL schemes to approximate the solutions of hyperbolic system of conservation laws. In the context of the 2D unstructured grids, we propose a limitation procedure on the gradient reconstruction to enforce several stability properties. We establish that the MUSCL scheme preserves the invariant domains and satisfy a set of entropy inequalities. A conservation assumption is not useful in the present work to define the piecewise linear approximations and the proposed limitation can be understood as a systematic correction of the standard gradient reconstruction procedure. The numerical method is applied to the compressible Euler equations. The gradient reconstruction is performed using the characteristic variables. Several numerical tests exhibit stability and robustness of the scheme.

Nonlinear projection methods for multi-entropies Navier–Stokes systems

This paper is devoted to the numerical approximation of the compressible Navier-Stokes equations with several independent entropies. Various models for complex compressible materials typically enter the proposed framework. The striking novelty over the usual Navier-Stokes equations stems from the generic impossibility of recasting equivalently the present system in full conservation form. Classical finite volume methods are shown to grossly fail in the capture of viscous shock solutions that are of primary interest in the present work. To enforce for validity a set of generalized jump conditions that we introduce, we propose a systematic and effective correction procedure, the so-called nonlinear projection method, and prove that it preserves all the stability properties satisfied by suitable Godunov-type methods. Numerical experiments assess the relevance of the method when exhibiting approximate solutions in close agreement with exact solutions.

Late-time/stiff-relaxation asymptotic-preserving approximations of hyperbolic equations

We investigate the late-time asymptotic behavior of solutions to nonlinear hyperbolic systems of conservation laws containing stiff relaxation terms. First, we introduce a Chapman-Enskog-type asymptotic expansion and derive an effective system of equations describing the late-time/stiff relaxation singular limit. The structure of this new system is discussed and the role of a mathematical entropy is emphasized. Second, we propose a new finite volume discretization which, in late-time asymptotics, allows us to recover a discrete version of the same effective asymptotic system. This is achieved provided we suitably discretize the relaxation term in a way that depends on a matrix-valued free-parameter, chosen so that the desired asymptotic behavior is obtained. Our results are illustrated with several models of interest in continuum physics, and numerical experiments demonstrate the relevance of the proposed theory and numerical strategy.

Why many theories of shock waves are necessary: kinetic relations for non-conservative systems

For a class of nonconservative hyperbolic systems of partial differential equations endowed with a strictly convex mathematical entropy, we formulate the initial value problem by supplementing the equations with a kinetic relation prescribing the rate of entropy dissipation across shock waves. Our condition can be regarded as a generalization to nonconservative systems of a similar concept introduced by Abeyaratne, Knowles, and Truskinovsky for subsonic phase transitions and by LeFloch for nonclassical undercompressive shocks to nonlinear hyperbolic systems. The proposed kinetic relation for nonconservative systems turns out to be equivalent, for the class of systems under consideration at least, to Dal Maso, LeFloch, and Murats definition based on a prescribed family of Lipschitz continuous paths. In agreement with previous theories, the kinetic relation should be derived from a phase plane analysis of traveling solutions associated with an augmented version of the nonconservative system. We illustrate with several examples that nonconservative systems arising in the applications fit in our framework, and for a typical model of turbulent fluid dynamics, we provide a detailed analysis of the existence and properties of traveling waves which yields the corresponding kinetic function.

Why the MUSCL–Hancock Scheme is L 1 -stable

The finite volume methods are one of the most popular numerical procedure to approximate the weak solutions of hyperbolic systems of conservation laws. They are developed in the framework of first-order numerical schemes. Several approaches are proposed to increase the order of accuracy. The van Leer methods are interesting ways. One of them, namely the MUSCL–Hancock scheme, is full time and space second-order accuracy. In the present work, we exhibit relevant conditions to ensure the L1-stability of the method. A CFL like condition is established, and a suitable limitation procedure for the gradient reconstruction is developed in order to satisfy the stability criterion. In addition, we show that the conservative variables are not useful within the gradient reconstruction and the procedure is extended in the framework of the primitive variables. Numerical experiments are performed to show the interest and the robustness of the method.