direct solver adjoint advection?
******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello, does the direct solver support adjoint advection: <I PROPERTY="SolverType" VALUE="CoupledLinearisedNS" /> <I PROPERTY="EQTYPE" VALUE="SteadyLinearisedNS" /> <I PROPERTY="EvolutionOperator" VALUE="Adjoint" /> Alex Proskurin Altai state technical university
Hi Alex, The direct solver solves the global linear system in the spectral element problem, like A x = b. In the incompressible flow solver, the nonlinear advection term is treated as a part of the right-hand-side term b. So, the direct solver can be used in adjoint advection problems. Cheers Ankang ________________________________ From: nektar-users-bounces@imperial.ac.uk <nektar-users-bounces@imperial.ac.uk> on behalf of A.P. <k210@list.ru> Sent: 31 July 2021 16:45 To: nektar-users <nektar-users@imperial.ac.uk> Subject: [Nektar-users] direct solver adjoint advection? This email from k210@list.ru originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list<https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address. Hello, does the direct solver support adjoint advection: <I PROPERTY="SolverType" VALUE="CoupledLinearisedNS" /> <I PROPERTY="EQTYPE" VALUE="SteadyLinearisedNS" /> <I PROPERTY="EvolutionOperator" VALUE="Adjoint" /> Alex Proskurin Altai state technical university
Hi Alex, I am sorry. I misunderstood your question yesterday. The direct solver I mentioned is for the "GlobalSysSoln" property, such as <I PROPERTY="GlobalSysSoln" VALUE="DirectMultiLevelStaticCond" /> <I PROPERTY="GlobalSysSoln" VALUE="XxtMultiLevelStaticCond" /> Its counterpart is the iterative solver, like "IterativeStaticCond". The direct solver for GlobalSysSoln is for general spectral element problems. The direct solver you said is the "CoupledLinearNS" for the incompressible N-S equation, and its counterpart is the VelocityCorrectionScheme. In the CoupledLinearNS solver, if you use EQTYPE=SteadyLinearisedNS, the equation to be solved is the linearized N-S equation, not its adjoint. The baseflow is defined by <FUNCTION NAME="AdvectionVelocity">. In this situation, the EvolutionOperator = Nonlinear,Direct,Adjoint,SkewSymmetric have the same effect. The adjoint equation of SteadyLinearisedNS has not been implemented in CoupledLinearNS yet. Cheers Ankang ________________________________ From: nektar-users-bounces@imperial.ac.uk <nektar-users-bounces@imperial.ac.uk> on behalf of Gao, Ankang <ankang.gao@imperial.ac.uk> Sent: 31 July 2021 19:17 To: nektar-users <nektar-users@imperial.ac.uk>; A.P. <k210@list.ru> Subject: Re: [Nektar-users] direct solver adjoint advection? Hi Alex, The direct solver solves the global linear system in the spectral element problem, like A x = b. In the incompressible flow solver, the nonlinear advection term is treated as a part of the right-hand-side term b. So, the direct solver can be used in adjoint advection problems. Cheers Ankang ________________________________ From: nektar-users-bounces@imperial.ac.uk <nektar-users-bounces@imperial.ac.uk> on behalf of A.P. <k210@list.ru> Sent: 31 July 2021 16:45 To: nektar-users <nektar-users@imperial.ac.uk> Subject: [Nektar-users] direct solver adjoint advection? This email from k210@list.ru originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list<https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address. Hello, does the direct solver support adjoint advection: <I PROPERTY="SolverType" VALUE="CoupledLinearisedNS" /> <I PROPERTY="EQTYPE" VALUE="SteadyLinearisedNS" /> <I PROPERTY="EvolutionOperator" VALUE="Adjoint" /> Alex Proskurin Altai state technical university
participants (2)
-
A.P.
-
Gao, Ankang