Compressible solver queries
Hi all, I was giving the compressible flow solver a try and have a couple of queries. 1. Are the history probes set up for this solver? 2. I was looking into the possibility of adding a passive scalar term for some validation. From a first glance, it seems that an additional scalar transport equation is not implemented in the CompressibleFlowSolver (at least in the v4.5.0). Is that correct? If yes, has anyone tried that yet? If not, could you please give me some ideas on how to start about it? I guess, we just need to add convective and diffusive fluxes to the system of equations somehow (in CompressibleFlowSystem.cpp and NavierStokesCFE.cpp?). And I guess the Boundary Conditions can be generically defined for extra variables as in CompressibleFlowSystem::DoOdeProjection function(?) PS. What is the "smooth viscosity model"? 446 // For the smooth viscosity model 447 if (nVariables == m_spacedim+3) 448 { 449 // Add a zero row for the advective fluxes 450 for (j = 0; j < m_spacedim; ++j) 451 { 452 Vmath::Zero(nq, flux[m_spacedim+2][j], 1); 453 } 454 } Thank you in advance for your time. Best Regards, Vishal --- *Vishal SAINI* PhD Student Loughborough University
Hi Vishal, 1. As far as I know the probes should work with any solver. 2. There are no passive scalars in the CompressibleFlowSolver, but you seem on the right track to start. I would start by duplicating the CompressibleFlowSystem (or better deriving a class from it, see for instance NavierStokesCFE.cpp) and adding the inviscid fluxes for the passive scalar/s and related BCs. The smooth viscosity model is the artificial viscosity model proposed by Barther and Darmofal in 2010. It is not working in its current implementation and we are not planning to fix it. I hope this helps. Cheers, Giacomo On Tue, Oct 8, 2019 at 11:25 AM Vishal Saini <vishal.saini.nitj@gmail.com> wrote:
Hi all,
I was giving the compressible flow solver a try and have a couple of queries.
1. Are the history probes set up for this solver?
2. I was looking into the possibility of adding a passive scalar term for some validation. From a first glance, it seems that an additional scalar transport equation is not implemented in the CompressibleFlowSolver (at least in the v4.5.0). Is that correct? If yes, has anyone tried that yet?
If not, could you please give me some ideas on how to start about it? I guess, we just need to add convective and diffusive fluxes to the system of equations somehow (in CompressibleFlowSystem.cpp and NavierStokesCFE.cpp?). And I guess the Boundary Conditions can be generically defined for extra variables as in CompressibleFlowSystem::DoOdeProjection function(?)
PS. What is the "smooth viscosity model"?
446 // For the smooth viscosity model 447 if (nVariables == m_spacedim+3) 448 { 449 // Add a zero row for the advective fluxes 450 for (j = 0; j < m_spacedim; ++j) 451 { 452 Vmath::Zero(nq, flux[m_spacedim+2][j], 1); 453 } 454 }
Thank you in advance for your time.
Best Regards, Vishal --- *Vishal SAINI* PhD Student Loughborough University _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
Hello Vishal, There are many places you will need to modify: CompressibleFlowSolver in equation systems, convective fluxes, Riemann solvers, boundary conditions, variable converter, diffusive fluxes, LDGNS (with it's boundary conditions). I have done on Nektar++ v4.4.3. I may share the code if you are interested. Aidyn On Tue, Oct 8, 2019, 06:25 Vishal Saini <vishal.saini.nitj@gmail.com> wrote:
Hi all,
I was giving the compressible flow solver a try and have a couple of queries.
1. Are the history probes set up for this solver?
2. I was looking into the possibility of adding a passive scalar term for some validation. From a first glance, it seems that an additional scalar transport equation is not implemented in the CompressibleFlowSolver (at least in the v4.5.0). Is that correct? If yes, has anyone tried that yet?
If not, could you please give me some ideas on how to start about it? I guess, we just need to add convective and diffusive fluxes to the system of equations somehow (in CompressibleFlowSystem.cpp and NavierStokesCFE.cpp?). And I guess the Boundary Conditions can be generically defined for extra variables as in CompressibleFlowSystem::DoOdeProjection function(?)
PS. What is the "smooth viscosity model"?
446 // For the smooth viscosity model 447 if (nVariables == m_spacedim+3) 448 { 449 // Add a zero row for the advective fluxes 450 for (j = 0; j < m_spacedim; ++j) 451 { 452 Vmath::Zero(nq, flux[m_spacedim+2][j], 1); 453 } 454 }
Thank you in advance for your time.
Best Regards, Vishal --- *Vishal SAINI* PhD Student Loughborough University _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
participants (3)
-
Aidyn Aitzhan
-
Giacomo Castiglioni
-
Vishal Saini