Grid refinement #838
Grid refinement #838
Commits on Oct 10, 2022
-
Created new Gridap Refinement module
This module will contain new abstractions and tools to implement mesh refinement strategies in Gridap.
Commits on Oct 13, 2022
Commits on Oct 17, 2022
Commits on Oct 20, 2022
Commits on Oct 23, 2022
-
RefinedTriangulation constructor now has checks.
We check conditions for parametric sizes Dc, Dp so that they are consistent with the model the triangulation originates from.
-
Commits on Oct 24, 2022
-
-
RefinedDiscreteModel now enforces its background model not to be of t…
…ype RefinedModel.
-
Commits on Oct 25, 2022
-
-
-
-
-
-
-
-
Added abstract type GridTransferOperators
- Added new abstract type GridTranferOperator representing an operator that moves data between two Gridap triangulations that are related by some refinement process. - Renamed RefinementTransferOperator as ProjectionTransferOperator, and rebranded it as a subtype of GridTransferOperator.
-
-
-
Commits on Oct 26, 2022
Commits on Oct 27, 2022
Commits on Oct 28, 2022
Commits on Nov 7, 2022
Commits on Nov 16, 2022
-
Measure is now abstract 2 include CompositeMeasure
- Measure renamed GenericMeasure - Created abstract type Measure - Moved CompositeMeasure to DomainContributions.jl
Commits on Nov 17, 2022
-
change_domain now does dispatch with source trian.
- Added strian=get_triangulation(a) argument to the change_domain call. This allows to do dispatch based on the type of source trian. - Added routines to get relationships between Adapted models and triangulations.
-
Commits on Nov 21, 2022
-
-
Started generalising for coarsening+refineing
- Created RefinementRules - New trait for AdaptivityGlue which allows to implement specialized routines for refinement-only cases (efficiency). - Refactored fine to coarse ~> old to new
-
-
Commits on Nov 22, 2022
Commits on Nov 24, 2022
Commits on Nov 28, 2022
Commits on Dec 2, 2022
Commits on Dec 5, 2022
Commits on Dec 6, 2022
-
-
-
-
Refactored refinement for cartesian meshes.
- Now works for arbitrary number of dimensions. - Now works for arbitrary number of refinements in each direction (i,e anisotropic refinement)
-
-
Merge branch 'refined-discrete-models' of github.com:gridap/Gridap.jl…
… into refined-discrete-models
Commits on Dec 7, 2022
Commits on Dec 8, 2022
-
Merge branch 'refined-discrete-models' of github.com:gridap/Gridap.jl…
… into refined-discrete-models
Commits on Dec 10, 2022
-
function refine(model::CartesianDiscreteModel{Dc}, cell_partition::Tuple) where Dc when Dc!=2. Why werent we detecting this bug with the current test suite?
Commits on Dec 20, 2022
Commits on Dec 22, 2022
-
Bugfix in fine->coarse when coarse trian was AdaptedTriangulation. Ad…
…ded tests for this case.
Commits on Dec 23, 2022
-
Merge branch 'refined-discrete-models' of github.com:gridap/Gridap.jl…
… into refined-discrete-models
Commits on Jan 5, 2023
-
-
-
AdaptivityGlue.refinement_rules is now a vector of RefinementRules defined on the old cells. This makes more sense when considering refinement-only scenarios and also enhances performance due to how reorderings interact with CompressedArrays/FillArrays.
-
Commits on Jan 6, 2023
Commits on Jan 8, 2023
-
-
Merge branch 'fine-to-coarse-fields-fast-evaluation' of github.com:gr…
…idap/Gridap.jl into refined-discrete-models
Commits on Jan 13, 2023
-
RefinementRule typing now homogeneous for mixed.
In order to work with lazy arrays, we need cell-arrays of RefinementRules to have a concrete and homogeneous typing. For meshes with mixed refining, like edge-based refined meshes, that was not the case. This is now fixed. Now arrays of FineToCoarseFields and FineToCoarseReferenceFEs have concrete eltype as well, despite containing a RefinementRule.
-