ForkMaster/netgen
1
0
Fork 0
mirror of https://github.com/NGSolve/netgen.git synced 2024-11-15 10:28:34 +05:00
Code Issues Packages Projects Releases Wiki Activity
4,750 Commits 10 Branches 43 Tags 18 MiB
dbffb79550
Commit Graph

56 Commits

Author SHA1 Message Date
Matthias Hochsteger
12f42b30cd New test results 2024-10-01 14:16:15 +02:00
Hochsteger, Matthias
9f6c64a4f9 Fix splitimprove edge 2024-09-27 01:00:50 +02:00
Hochsteger, Matthias
73c75240f8 Skip SplitImproveEdge if it would insert tets with negative volume 2024-09-13 10:22:50 +02:00
Hochsteger, Matthias
d0ba2934df Consistent penalty for illegal tets 2024-03-11 21:10:56 +01:00
Matthias Hochsteger
13867ef1a0 Extra optimization steps for bad elements 2024-02-28 10:07:57 +01:00
Matthias Hochsteger
0d481b1104 Use badness stored in Element 2024-02-23 18:20:14 +01:00
Matthias Hochsteger
e5b544e02b Don't optimize when goal==OPT_LEGAL and all affected elements are legal 2024-02-23 17:01:40 +01:00
Matthias Hochsteger
9cc6fc933d evalute localh in multiple points and take average 
- also apply combine improve (2d) in both directions
- new test results
 2022-04-13 12:44:45 +02:00
mhochsteger@cerbsim.com
d2348a7651 optimize occ surfaces individually again, update test results 2022-03-02 15:42:44 +01:00
mhochsteger@cerbsim.com
0f6dd37aa2 new meshing results 2022-02-23 21:23:41 +01:00
mhochsteger@cerbsim.com
a68cd9f89a fix parallel surface optimization 
update test results
 2022-02-15 19:09:53 +01:00
mhochsteger@cerbsim.com
45bd63810a update test results 2021-11-30 19:43:32 +01:00
Joachim Schoeberl
56e848eea9 pywrapping occ 2021-07-24 19:53:30 +02:00
Joachim Schoeberl
057d6a4ed4 change number of volelements due to different surf-meshing 2021-07-24 16:33:39 +02:00
Joachim Schoeberl
c543b03401 change number of surf elements due to different surf-meshing 2021-07-24 16:27:10 +02:00
Matthias Hochsteger
2231c06efb new results 2021-07-19 12:22:18 +02:00
Matthias Hochsteger
8baccf0a08 some fixes, new test results 2021-06-28 17:48:21 +02:00
Matthias Hochsteger
479efea50f test results 2021-06-16 15:25:28 +02:00
Matthias Hochsteger
9488485f22 update results 2021-06-14 10:39:57 +02:00
Matthias Hochsteger
97623db219 Mesh 3d domains in parallel 
To get consistent results, copy the LocalH tree in BlockFillLocalH
 2021-06-14 10:39:57 +02:00
Matthias Hochsteger
5e3505b897 faster SwapImprove after delaunay 
also new test results
 2021-06-14 10:39:57 +02:00
Matthias Hochsteger
2a17451c96 consistent criterion for conformance optimization in SwapImprove 
Only small effect on output Meshes
 2021-06-10 11:04:41 +02:00
Matthias Hochsteger
f7a08df155 Use sum of badnesses and not average value in CombineImprove in 2D 2021-05-14 17:55:09 +02:00
Matthias Hochsteger
7baea9801e adapt test results ( one test case removed ) 2021-05-14 17:39:06 +02:00
Christopher Lackner
4e2d2943f6 fix csg extrusion 2021-04-13 12:11:10 +02:00
Matthias Hochsteger
d7d12ac53d Don't swap very bad elements in SwapImprove 
Changes meshing -> new test results
 2021-03-01 10:08:00 +01:00
Matthias Hochsteger
fbeb6137eb Fix SwapImprove2 
Don't allow swaps if an adjacent element was deleted in the current
optimization pass. Also update test restults.
 2020-11-25 17:55:44 +01:00
Matthias Hochsteger
b28a1bc5cc update test results 2020-10-19 15:45:16 +02:00
Matthias Hochsteger
dab18a1c8a optimize CalcPartition() and PartitionBoundary() 
new test results (2d meshing behaviour is changed with this commit)
 2020-10-16 18:48:22 +02:00
Matthias Hochsteger
c0b8b1c0cc Parallel SplitImprove2, update test results 
Due to prallelization, the order of splits is changed (sort by
    improvement of badness, like in other optimization passes)
 2020-07-27 14:26:45 +02:00
Matthias Hochsteger
b689d13efe SplitImprove2 - optimize only bad elements, update results 2020-07-23 13:18:14 +02:00
Matthias Hochsteger
ce8ba71f33 Fix SwapImprove 2020-07-15 17:26:39 +00:00
Matthias Hochsteger
2faf16e8cf Update test results 2019-11-19 09:54:00 +01:00
Matthias Hochsteger
9ebc6b00f3 [testing] update results 2019-10-31 15:38:40 +00:00
Matthias Hochsteger
3acadc023f Rework on 3D SplitImprove 
- Try (expensive) split operation only if badness > 100 (of any adjacent
        tet) and badness > 0.1 * badness_max
- Use FindInnerPoint before BFGS to ensure valid starting point (similar
        as in ImproveMesh() )
- Skip valid tets if goal==OPT_LEGAL (as in sequential version)
 2019-10-29 17:58:37 +01:00
Matthias Hochsteger
2a51bf76c1 [testing] Add results for OCC 7.4, skip tests on Ubuntu (with OCC 7.3) 2019-10-28 15:11:42 +01:00
Matthias Hochsteger
63aab9076c parallel SplitImprove 2019-10-28 09:38:53 +01:00
Matthias Hochsteger
ab024c2e6c Sort edges by improvement in SwapImprove2 2019-10-24 18:04:56 +02:00
Matthias Hochsteger
6fb8f90a73 Sort edges by improvement in SwapImprove 2019-10-24 17:40:20 +02:00
Matthias Hochsteger
ffcf2c7373 [testing] Check for min/max trig/tet angles 2019-10-24 10:02:00 +02:00
Matthias Hochsteger
91815955f5 Test in2d files, update results 2019-10-23 13:44:33 +02:00
Matthias Hochsteger
c98ecafa2f Update results 2019-10-18 14:27:34 +02:00
Matthias Hochsteger
7c96e22a60 Remove one test case with inconsistent results 2019-10-14 17:59:45 +02:00
Matthias Hochsteger
6ca6a5e791 CombineImprove() 2D: optimize all faces at once 2019-10-14 16:21:24 +02:00
Matthias Hochsteger
0f095281d9 Parallel 2d CombineImprove() (also sort by improvement) 2019-10-14 16:02:11 +02:00
Matthias Hochsteger
294fbb0e6f Loop over edges in CombineImprove() 2019-10-14 14:04:10 +02:00
Matthias Hochsteger
59c355dbed New Table for elementsonnode 2019-10-14 11:49:23 +02:00
Matthias Hochsteger
268f2466f0 Parallel 2d MeshImprove 2019-10-10 15:44:57 +02:00
Matthias Hochsteger
18bdd9df93 Build neighbors list in parallel 
Check in new results (part1.stl does not mesh with very_coarse anymore)
 2019-10-08 19:26:43 +02:00
Matthias Hochsteger
f24a749fb2 Parallel MeshOptimize2d::EdgeSwapping() (individual faces) 2019-10-08 17:29:38 +02:00