This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| HLLC | |
|---|---|
| Name | HLLC |
| Field | Computational Fluid Dynamics |
| Introduced | 1993 |
| Authors | Einfeldt, Harten, Lax, van Leer, Toro |
| Classification | Riemann solver, approximate |
| Related | Roe solver, HLL, Godunov method |
HLLC
HLLC is an approximate Riemann solver used in numerical solutions of hyperbolic conservation laws, developed to capture contact and shear waves missing in earlier two-wave approximations. It builds on work by Riemann, Einfeldt, Harten, Lax, and van Leer and is often paired with high-resolution schemes by Godunov, TVD, and MUSCL methods. Widely applied in codes developed by groups at NASA, CERN, Los Alamos National Laboratory, and universities such as MIT, Stanford University, and University of Cambridge, HLLC has influenced solvers implemented in libraries like OpenFOAM, FLASH (software), Athena (software), and PLUTO (code). Practitioners in aerospace projects at Boeing, Airbus, and institutes like JAXA and ESA rely on HLLC for compressible flow modeling in research appearing in journals like Journal of Computational Physics and AIAA Journal.
The HLLC formulation approximates the solution of the one-dimensional Riemann problem for conservation laws introduced by Riemann and extended in the framework of finite-volume methods by Godunov. It assumes three constant states separated by two nonlinear waves and a middle contact discontinuity analogous to constructions by Toro (Paolo) and Einfeldt (S.). Starting from integral conservation across characteristic curves used by Lax, the solver expresses fluxes via estimated wave speeds often taken from methods of Davis (S.), Roe (P.), and Liou (H.). The Rankine–Hugoniot conditions named after Rankine and Hugoniot enforce jump relations across shocks; HLLC augments the HLL two-wave model of Harten (A.), Lax (P.D.), and van Leer (B.) by restoring the contact wave first emphasized in studies by Batchelor, Courant, and Friedrichs (K.O.). The resulting flux function uses left and right conservative state vectors as in schemes derived by LeVeque and employs eigenstructure considerations related to analyses by Godunov (S.K.) and Roe.
Implementations of HLLC integrate with reconstruction strategies from ENO and WENO families developed by Harten, Engquist, and Osher (S.), or with piecewise-linear slope limiting following van Leer and Monotone Upstream-centered Schemes for Conservation Laws (MUSCL) popularized by van Albada and Minmod limiters. Time integration commonly uses Runge–Kutta schemes studied by Shu (C.-W.) and Osher or multi-stage methods from Butcher (J.C.). Practical codes require robust wave-speed estimates such as those proposed by Davis (S.), Einfeldt (B.), and Toro, and may incorporate positivity-preserving strategies inspired by Shu and Zhang (X.). Boundary condition treatments draw on methodologies from Courant (R.), Friedrichs, and implementations in community codes like OpenFOAM and FLASH (software). Parallelization for distributed memory machines follows paradigms established at Argonne National Laboratory, Lawrence Livermore National Laboratory, and Los Alamos National Laboratory.
Extensions of the base HLLC solver include adaptations for magnetohydrodynamics influenced by the work of Brio and Wu and constrained transport techniques associated with Evans (C.) and Hawley (J.F.). Multi-dimensional generalizations employ unsplit integrators like those by Colella (P.) and Saltzman and transverse Riemann solvers following LeVeque. Low-dissipation and entropy-fix modifications echo approaches of Roe and Harten, while positivity-preserving and well-balanced variants leverage ideas from Greenberg, LeVeque, and Audusse (E.). Hybrid schemes combine HLLC with approximate solvers such as Rusanov, AUSM, and Roe-type fluxes used in work from Jameson (A.) and Venkateswaran.
HLLC is employed in a broad spectrum of computational studies: supersonic and hypersonic aerodynamics in projects at NASA and ESA, astrophysical simulations of supernovae and accretion disks by teams at Max Planck Institute for Astrophysics, Caltech, and Princeton University, and inertial confinement fusion modeling at Lawrence Livermore National Laboratory. It is used in weather and climate components within models developed by NOAA and Met Office groups for shock-driven flows, in industrial CFD for turbomachinery at Siemens and General Electric, and in research on relativistic jets by collaborations involving Rutgers University and Columbia University.
Benchmark studies compare HLLC against Roe, HLL, and exact Riemann solvers in test suites originating from Sod (G.A.), Lax, and Woodward (P.); results reported in Journal of Computational Physics and AIAA Journal show HLLC balances accuracy and robustness, resolving contacts better than HLL while avoiding some nonphysical states that Roe sometimes induces. In magnetohydrodynamics, HLLC-based extensions compete with HLLD solvers developed by Miyoshi (T.) and Kudoh (H.), trading complexity for stability. Performance on modern hardware follows optimization strategies from Intel, NVIDIA, and AMD for vectorization and GPU acceleration implemented in community codes like Athena (software) and PLUTO (code).
The HLLC solver emerged in the early 1990s as an evolution of ideas by Harten, Lax, and van Leer with concrete formulations credited to researchers including Einfeldt and Toro (Paolo). Subsequent contributors refined wave-speed estimates and extended the method to magnetohydrodynamics and relativistic flows through work at institutions such as University of Michigan, University of Illinois Urbana–Champaign, Imperial College London, and national laboratories including Los Alamos National Laboratory and Lawrence Livermore National Laboratory. The solver’s adoption grew via integration into open-source projects like OpenFOAM and widely cited textbooks by Toro (Paolo), which disseminated practical implementations to academia and industry.
Category:Numerical methods