feat: make possible to use solid mechanics solver to perform poromechanics initialization #3396
Conversation
paveltomin
requested review from
rrsettgast,
castelletto1,
CusiniM,
klevzoff,
francoishamon,
cssherman and
jhuang2601
as code owners
paveltomin
added
the
ci: run integrated tests
inputFiles/poromechanics/PoroElastic_staircase_co2_3d_fim_new.xml
Outdated
Show resolved
Hide resolved
| @@ -118,6 +114,10 @@ class PoromechanicsSolver : public CoupledSolver< FLOW_SOLVER, MECHANICS_SOLVER | |||
| GEOS_FMT( "{} {}: The attribute `{}` of the flow solver `{}` must be set to 1 since the poromechanics solver is thermal", | |||
| this->getCatalogName(), this->getName(), FlowSolverBase::viewKeyStruct::isThermalString(), this->flowSolver()->getName() ), | |||
| InputError ); | |||
|
|
|||
| DomainPartition & domain = this->template getGroupByPath< DomainPartition >( "/Problem/domain" ); | |||
| flowSolver()->initialize( domain ); | |||
There was a problem hiding this comment.
need to call this first to initialize porosity and densities
There was a problem hiding this comment.
I'm surprised that initialization of these quantities can be called from a constructor.... versus in some post process routine... it's good to have explicit naming for initialization... I'm assuming this is capillary gravity equilibrium for the reservoir solver ?
There was a problem hiding this comment.
it's a bunch of things including hydrostatic equilibrium
multiphase equilibration is not yet supported, see
src/coreComponents/physicsSolvers/multiphysics/PoromechanicsSolver.hpp
Outdated
Show resolved
Hide resolved
paveltomin
changed the title
| @@ -1055,7 +1056,7 @@ void SolidMechanicsLagrangianFEM::assembleSystem( real64 const GEOS_UNUSED_PARAM | |||
| MeshLevel & mesh, | |||
| arrayView1d< string const > const & regionNames ) | |||
| { | |||
| if( m_isFixedStressPoromechanicsUpdate ) | |||
| if( m_isFixedStressPoromechanicsUpdate || m_performStressInitialization ) | |||
There was a problem hiding this comment.
Not that this PR is the introduction of this, but the single physics solver shouldn't know about the coupled physics solver.
There was a problem hiding this comment.
In the case of coupled reservoir and well solver, where the event section only activates the reservoir solver, doesn't the well solver need to know that reservoir solver is assuming that well variables are constant. the well solver knows its coupled to the reservoir via perforations.
|
@paveltomin Why can the mechanics solver not be called directly from the poromechanics initialization? I am missing something. |
I am not sure I understood you correctly, sorry. |
paveltomin
added
the
ci: run integrated tests
paveltomin
removed request for
klevzoff,
francoishamon,
cssherman and
wrtobin
paveltomin
added
the
flag: requires rebaseline
|
new baselines are needed but no real results changes |
paveltomin
removed
the
ci: run integrated tests
and call it with poromechanics solver pointer except for hydrofrac case