Raphaël Loubère

ReALE: A reconnection-based arbitrary-Lagrangian-Eulerian method

Abstract We present a new reconnection-based arbitrary-Lagrangian–Eulerian (ALE) method. The main elements in a standard ALE simulation are an explicit Lagrangian phase in which the solution and grid are updated, a rezoning phase in which a new grid is defined, and a remapping phase in which the Lagrangian solution is transferred (conservatively interpolated) onto the new grid. In standard ALE methods the new mesh from the rezone phase is obtained by moving grid nodes without changing connectivity of the mesh. Such rezone strategy has its limitation due to the fixed topology of the mesh. In our new method we allow connectivity of the mesh to change in rezone phase, which leads to general polygonal mesh and allows to follow Lagrangian features of the mesh much better than for standard ALE methods. Rezone strategy with reconnection is based on using Voronoi tessellation. We demonstrate performance of our new method on series of numerical examples and show it superiority in comparison with standard ALE methods without reconnection.

A posteriori subcell limiting of the discontinuous Galerkin finite element method for hyperbolic conservation laws

The purpose of this work is to propose a novel a posteriori finite volume subcell limiter technique for the Discontinuous Galerkin finite element method for nonlinear systems of hyperbolic conservation laws in multiple space dimensions that works well for arbitrary high order of accuracy in space and time and that does not destroy the natural subcell resolution properties of the DG method. High order time discretization is achieved via a one-step ADER approach that uses a local space-time discontinuous Galerkin predictor method to evolve the data locally in time within each cell.Our new limiting strategy is based on the so-called MOOD paradigm, which a posteriori verifies the validity of a discrete candidate solution against physical and numerical detection criteria after each time step. Here, we employ a relaxed discrete maximum principle in the sense of piecewise polynomials and the positivity of the numerical solution as detection criteria. Within the DG scheme on the main grid, the discrete solution is represented by piecewise polynomials of degree N. For those troubled cells that need limiting, our new limiter approach recomputes the discrete solution by scattering the DG polynomials at the previous time step onto a set of N s = 2 N + 1 finite volume subcells per space dimension. A robust but accurate ADER-WENO finite volume scheme then updates the subcell averages of the conservative variables within the detected troubled cells. The recomputed subcell averages are subsequently gathered back into high order cell-centered DG polynomials on the main grid via a subgrid reconstruction operator. The choice of N s = 2 N + 1 subcells is optimal since it allows to match the maximum admissible time step of the finite volume scheme on the subgrid with the maximum admissible time step of the DG scheme on the main grid, minimizing at the same time also the local truncation error of the subcell finite volume scheme. It furthermore provides an excellent subcell resolution of discontinuities.Our new approach is therefore radically different from classical DG limiters, where the limiter is using TVB or (H)WENO reconstruction based on the discrete solution of the DG scheme on the main grid at the new time level. In our case, the discrete solution is recomputed within the troubled cells from the old time level using a different and more robust numerical scheme on a subgrid level.We illustrate the performance of the new a posteriori subcell ADER-WENO finite volume limiter approach for very high order DG methods via the simulation of numerous test cases run on Cartesian grids in two and three space dimensions, using DG schemes of up to tenth order of accuracy in space and time ( N = 9 ). The method is also able to run on massively parallel large scale supercomputing infrastructure, which is shown via one 3D test problem that uses 10 billion space-time degrees of freedom per time step. A posteriori subcell limiter for arbitrary high order DG schemes in space and time.Simple a posteriori troubled zones indicator.The new limiter maintains the DG subgrid resolution capability even at discontinuities.Troubled zones are recomputed with ADER-WENO finite volume scheme on the subgrid.HPC simulations in 3D with P9 elements and 10 billion degrees of freedom.

A nominally second-order cell-centered Lagrangian scheme for simulating elastic-plastic flows on two-dimensional unstructured grids

In this paper, we describe a cell-centered Lagrangian scheme devoted to the numerical simulation of solid dynamics on two-dimensional unstructured grids in planar geometry. This numerical method, utilizes the classical elastic-perfectly plastic material model initially proposed by Wilkins [M.L. Wilkins, Calculation of elastic-plastic flow, Meth. Comput. Phys. (1964)]. In this model, the Cauchy stress tensor is decomposed into the sum of its deviatoric part and the thermodynamic pressure which is defined by means of an equation of state. Regarding the deviatoric stress, its time evolution is governed by a classical constitutive law for isotropic material. The plasticity model employs the von Mises yield criterion and is implemented by means of the radial return algorithm. The numerical scheme relies on a finite volume cell-centered method wherein numerical fluxes are expressed in terms of sub-cell force. The generic form of the sub-cell force is obtained by requiring the scheme to satisfy a semi-discrete dissipation inequality. Sub-cell force and nodal velocity to move the grid are computed consistently with cell volume variation by means of a node-centered solver, which results from total energy conservation. The nominally second-order extension is achieved by developing a two-dimensional extension in the Lagrangian framework of the Generalized Riemann Problem methodology, introduced by Ben-Artzi and Falcovitz [M. Ben-Artzi, J. Falcovitz, Generalized Riemann Problems in Computational Fluid Dynamics, Cambridge Monogr. Appl. Comput. Math. (2003)]. Finally, the robustness and the accuracy of the numerical scheme are assessed through the computation of several test cases.

The internal consistency, stability, and accuracy of the discrete, compatible formulation of Lagrangian hydrodynamics

This work explores the somewhat subtle meaning and consequences of the salient properties of the discrete, compatible formulation of Lagrangian hydrodynamics. In particular, since this formulation preserves total energy to roundoff error, the amount of error in the conservation of total energy cannot be used to gauge the internal consistency of calculations, as is often done with the older forms of this algorithm. However, the compatible formulation utilizes two definitions of zone volume: the first is the usual definition whereby the volume of a zone is defined as some prescribed function of the coordinates of the points that define it; the second is given as the integration in time of the continuity equation for zone volume as expressed in Lagrangian form. It is the use of this latter volume in the specific internal energy equation that enables total energy to be exactly conserved. These two volume definitions are generally not precisely equal. It is the analysis of this difference that forms the first part of this study. It is shown that this difference in zone volumes can be used to construct a practical internal consistency measure that not only takes the place of the lack of total energy conservation of the older forms of Lagrangian hydrodynamics, but is more general in that it can be defined on a single zone basis. It can also be used to ascertain the underlying spatial and temporal order of accuracy of any given set of calculations. The difference in these two definitions of zone volume may be interpreted as a type of entropy error. However, this entropy error is found to be significant only when a given calculation becomes numerically unstable, otherwise it remains at or far beneath truncation error levels. In fact, it can be utilized to provide an upper bound on the size of the spatial truncation error for a stable computation. It is also shown how this volume difference can be used as an indicator of numerical difficulties, since exact local conservation of total energy does not guarantee numerical stability or the quality of any numerical calculation. The discrete, compatible formulation of Lagrangian hydrodynamics utilizes a two level predictor/corrector-type of time integration scheme; a stability analysis, both analytical and numerical, is given. This analysis reveals a novel stability diagram that has not been heretofore published, and gives definitive information as to how the stabilizing corrector step should be centered in time.

A second-order compatible staggered Lagrangian hydrodynamics scheme using a cell-centered multidimensional approximate Riemann solver

We develop a general framework to derive and analyze staggered numerical schemes devoted to solve hydrodynamics equations in 2D. In this framework a cell-centered multi-dimensional approximate Riemann solver is used to build a form of artificial viscosity that leads to a conservative, compatible and thermodynamically consistent scheme. A second order extension in space and time for this scheme is proposed in this work and we prove on numerical examples the validity of this approach.

A second-order accurate material-order-independent interface reconstruction technique for multi-material flow simulations

A new, second-order accurate, volume conservative, material-order-independent interface reconstruction method for multi-material flow simulations is presented. First, materials are located in multi-material computational cells using a piecewise linear reconstruction of the volume fraction function. These material locator points are then used as generators to reconstruct the interface with a weighted Voronoi diagram that matches the volume fractions. The interfaces are then improved by minimizing an objective function that smoothes interface normals while enforcing convexity and volume constraints for the pure material subcells. Convergence tests are shown demonstrating second-order accuracy. Static and dynamic examples are shown illustrating the superior performance of the method over existing material-order-dependent methods.

Direct Arbitrary-Lagrangian-Eulerian ADER-MOOD finite volume schemes for multidimensional hyperbolic conservation laws

In this paper we present a new family of efficient high order accurate direct Arbitrary-Lagrangian-Eulerian (ALE) one-step ADER-MOOD finite volume schemes for the solution of nonlinear hyperbolic systems of conservation laws for moving unstructured triangular and tetrahedral meshes. This family is the next generation of the ALE ADER-WENO schemes presented in 16,20]. Here, we use again an element-local space-time Galerkin finite element predictor method to achieve a high order accurate one-step time discretization, while the somewhat expensive WENO approach on moving meshes, used to obtain high order of accuracy in space, is replaced by an a posteriori MOOD loop which is shown to be less expensive but still as accurate. This a posteriori MOOD loop ensures the numerical solution in each cell at any discrete time level to fulfill a set of user-defined detection criteria. If a cell average does not satisfy the detection criteria, then the solution is locally re-computed by progressively decrementing the order of the polynomial reconstruction, following a so-called cascade of predefined schemes with decreasing approximation order. A so-called parachute scheme, typically a very robust first order Godunov-type finite volume method, is employed as a last resort for highly problematic cells. The cascade of schemes defines how the decrementing process is carried out, i.e. how many schemes are tried and which orders are adopted for the polynomial reconstructions. The cascade and the parachute scheme are choices of the user or the code developer. Consequently the iterative MOOD loop allows the numerical solution to maintain some interesting properties such as positivity, mesh validity, etc., which are otherwise difficult to ensure. We have applied our new high order unstructured direct ALE ADER-MOOD schemes to the multi-dimensional Euler equations of compressible gas dynamics. A large set of test problems has been simulated and analyzed to assess the validity of our approach in terms of both accuracy and efficiency (CPU time and memory consumption).

A simple robust and accurate a posteriori sub-cell finite volume limiter for the discontinuous Galerkin method on unstructured meshes

In this paper we propose a simple, robust and accurate nonlinear a posteriori stabilization of the Discontinuous Galerkin (DG) finite element method for the solution of nonlinear hyperbolic PDE systems on unstructured triangular and tetrahedral meshes in two and three space dimensions. This novel a posteriori limiter, which has been recently proposed for the simple Cartesian grid case in 62, is able to resolve discontinuities at a sub-grid scale and is substantially extended here to general unstructured simplex meshes in 2D and 3D. It can be summarized as follows:At the beginning of each time step, an approximation of the local minimum and maximum of the discrete solution is computed for each cell, taking into account also the vertex neighbors of an element. Then, an unlimited discontinuous Galerkin scheme of approximation degree N is run for one time step to produce a so-called candidate solution. Subsequently, an a posteriori detection step checks the unlimited candidate solution at time t n + 1 for positivity, absence of floating point errors and whether the discrete solution has remained within or at least very close to the bounds given by the local minimum and maximum computed in the first step. Elements that do not satisfy all the previously mentioned detection criteria are flagged as troubled cells. For these troubled cells, the candidate solution is discarded as inappropriate and consequently needs to be recomputed. Within these troubled cells the old discrete solution at the previous time t n is scattered onto small sub-cells ( N s = 2 N + 1 sub-cells per element edge), in order to obtain a set of sub-cell averages at time t n . Then, a more robust second order TVD finite volume scheme is applied to update the sub-cell averages within the troubled DG cells from time t n to time t n + 1 . The new sub-grid data at time t n + 1 are finally gathered back into a valid cell-centered DG polynomial of degree N by using a classical conservative and higher order accurate finite volume reconstruction technique.Consequently, if the number N s is sufficiently large ( N s ź N + 1 ), the subscale resolution capability of the DG scheme is fully maintained, while preserving at the same time an essentially non-oscillatory behavior of the solution at discontinuities. Many standard DG limiters only adjust the discrete solution in troubled cells, based on the limiting of higher order moments or by applying a nonlinear WENO/HWENO reconstruction on the data at the new time t n + 1 . Instead, our new DG limiter entirely recomputes the troubled cells by solving the governing PDE system again starting from valid data at the old time level t n , but using this time a more robust scheme on the sub-grid level. In other words, the piecewise polynomials produced by the new limiter are the result of a more robust solution of the PDE system itself, while most standard DG limiters are simply based on a mere nonlinear data post-processing of the discrete solution. Technically speaking, the new method corresponds to an element-wise checkpointing and restarting of the solver, using a lower order scheme on the sub-grid. As a result, the present DG limiter is even able to cure floating point errors like NaN values that have occurred after divisions by zero or after the computation of roots from negative numbers. This is a unique feature of our new algorithm among existing DG limiters.The new a posteriori sub-cell stabilization approach is developed within a high order accurate one-step ADER-DG framework on multidimensional unstructured meshes for hyperbolic systems of conservation laws as well as for hyperbolic PDE with non-conservative products. The method is applied to the Euler equations of compressible gas dynamics, to the ideal magneto-hydrodynamics equations (MHD) as well as to the seven-equation Baer-Nunziato model of compressible multi-phase flows. A large set of standard test problems is solved in order to assess the accuracy and robustness of the new limiter. New sub-cell finite volume limiter for DG schemes on unstructured meshes in 2D and 3D.Simple a posteriori troubled zones indicator.Fully automatic sub-grid generation via nodes known from high order iso-parametric FEM.The new limiter maintains the DG sub-grid resolution property even at discontinuities.Element-wise checkpointing and restarting allows to cure floating point errors (NaN) at runtime.

Short Note: Volume consistency in a staggered grid Lagrangian hydrodynamics scheme

Staggered grid Lagrangian schemes for compressible hydrodynamics involve a choice of how internal energy is advanced in time. The options depend on two ways of defining cell volumes: an indirect one, that guarantees total energy conservation, and a direct one that computes the volume from its definition as a function of the cell vertices. It is shown that the motion of the vertices can be defined so that the two volume definitions are identical. A so modified total energy conserving staggered scheme is applied to the Coggeshall adiabatic compression problem, and now also entropy is basically exactly conserved for each Lagrangian cell, and there is increased accuracy for internal energy. The overall improvement as the grid is refined is less than what might be expected.

Towards an ultra efficient kinetic scheme. Part II

In this paper we present a new ultra efficient numerical method for solving kinetic equations. In this preliminary work, we present the scheme in the case of the BGK relaxation operator. The scheme, being based on a splitting technique between transport and collision, can be easily extended to other collisional operators as the Boltzmann collision integral or to other kinetic equations such as the Vlasov equation. The key idea, on which the method relies, is to solve the collision part on a grid and then to solve exactly the transport linear part by following the characteristics backward in time. The main difference between the method proposed and semi-Lagrangian methods is that here we do not need to reconstruct the distribution function at each time step. This allows to tremendously reduce the computational cost of the method and it permits for the first time, to the author@?s knowledge, to compute solutions of full six dimensional kinetic equations on a single processor laptop machine. Numerical examples, up to the full three dimensional case, are presented which validate the method and assess its efficiency in 1D, 2D and 3D.