Skip to content
Started 3 yr 9 mo ago
Took 11 hr on reynolds
Success

#264 (Jun 13, 2020, 1:52:07 AM)

Changes
  1. relocate QauxvInterpret forward declaration in prep for 3D stuff (details)
  2. Added basic stuff for 3D IBL (details)
  3. Minor cleanup (details)
  4. Initial pass on PDEIBL3D (details)
  5. Filled in 3D IBL closure with fully 3D components set to zero (details)
  6. Added BCNone for IBL3D (details)
  7. added a working version of sandbox case for IBL3D (details)
  8. Added a sheet grid of triangles in 3D for IBL3D (details)
  9. Added instantiation (details)
  10. Intermediate commit (details)
  11. Working on integrands for first 3D IBL case (details)
  12. More fixes to integrand classes for IBL3D (details)
  13. Fixed instantiations for IBL3D (details)
  14. Add explicit instantiations and fix typos in closure class (details)
  15. Initial version of local Cartesian basis construction for 3D IBL (details)
  16. Minor cleanup (details)
  17. Fix constuction of q1 and q2 vectors in case of stagnation flows (details)
  18. Add velocity gradient calculation in cell integrand of 3D IBL (details)
  19. Refactor and add a 2D stagnation flow test case (details)
  20. Add a 3D quad grid (details)
  21. Update cmake list for unit grid test (details)
  22. Fix typo in 3D quad sheet grid (details)
  23. fix bug in qauxv gradient calculation in 3D IBL cell integrand (details)
  24. Fix the scale and dimension of dissipation integral in 3D integral (details)
  25. fix a typo in thickness identity (details)
  26. Add more options to 3D IBL stagnation flow case, including quad grid, solution initialization, (details)
  27. one more bug fix in cell integrand for 3D IBL (details)
  28. Bugfix in calculation of trace unit normal vector for area cells in 3D (details)
  29. Add Nishida's crossflow closure.  Add pyrite check of stagnation flow case for 3D IBL.  Minor addition for debugging in PDEIBL3D (details)
  30. fix dependents of refactoring of GetSolnDofCell_IBL (details)
  31. Add a 3D flat plate source panel method but haven't tested yet (details)
  32. intermediate commit: minor clenaup and add guards. (details)
  33. intermediate commit: working on 3D coupled IBL for torpedo cass (details)
  34. add 3x3 equation system (details)
  35. typo fix. add partial instantiations for system 3x3 vector and matrix types (details)
  36. add linesearch data type for 3x3 equation system (details)
  37. add missing include (details)
  38. intermediate commit: fix many compiling errors in the way of running torpedo case for IBL3D. switching computer to continue the work... (details)
  39. refactor projection to Qauxi (details)
  40. add interior trace integrand for projection onto Qauxi (details)
  41. minor fix (details)
  42. add instantiations for IBL3D torpedo case (details)
  43. add instantiations for coupled IBL3D (details)
  44. remove unnecessary assert (details)
  45. fix for coupled IBL case to compile and run (details)
  46. fix uninitialized dummy DOFs (details)
  47. fix setDOFFrom in Qauxv 3D (missing q_y previously) (details)
  48. Coupled 3D IBL case converged on some simple variants of the torpedo case: linear and mildly nonlinear radial velocity distribution with attached laminar BLs (details)
  49. Tried triangle grid as well: fine on radially linear velocity distribution but couldn't converge on a more nonlinear case (details)
  50. cleaned up unused stuff (details)
  51. refactor SetIBLoutputCellGroup to accommodate both 3D and 2D IBL (details)
  52. remove outdated include (details)
  53. more refactoring of SetIBLoutputCellGroup for 3D IBL.  working progress.  switching computer (details)
  54. intermediate commit. switching computer. (details)
  55. enable tecplot output for 3D IBL (details)
  56. add full-state BC for IBL3D with corresponding boundary trace integrand.  verified on an uncoupled Blasius case (details)
  57. add a coupled IBL3D Blasius case (details)
  58. implement PTC for coupled IBL 3x3 system (details)
  59. add commented-out stuff for turning off laminar amplification and lag equations in IBL3D (details)
  60. fix typo in commented-out stuff (details)
  61. enable rotated mesh for IBL3D torpedo case (details)
  62. set up a torpedo with grid aligned with xy axes.  turned off laminar amplification and lag equations [need to be brought back after testing purely laminar BLs] (details)
  63. minor change to test case (details)
  64. minor refactoring on Blasius inflow BC params (details)
  65. rename test cases (details)
  66. speed up panel-related jacobian calculation using cached influence matrix in place of surrealized qauxifld (details)
  67. add jacobian timing for refactoring purpose (details)
  68. add similar (to DBL3) Falkner-Skan inflow bc for IBL3D with non-constant inflow state. set up a tornado case with rotated xy grid to compare with DBL3 (details)
  69. minor refactoring of test cases (details)
  70. remove trivial zero z velocity component to speed up panel method (details)
  71. minor refactoring (details)
  72. typo fix (details)
  73. precompute inviscid velocity for panel, which resulted in O(10) speedup in related residual calculation. fix DOF indexing in residual/jacobian precomputation (details)
  74. define macro to comment off some lines for timing/refactoring (details)
  75. minor comments (details)
  76. add ping test for coupled IBL in 3x3 block system (details)
  77. fix typo in Cf2 closure, although it didn't change the torpedo case results much (details)
  78. add a wall term to dissipation relation of Nishida's closure, although it didn't affect things much (details)
  79. add a test case of fx67-k-170 airfoil (details)
  80. minor edit to output filenames at convergence failures (details)
  81. resolve merge conflict (details)
  82. changed default flux of IBL to LF flux (details)
  83. add a test case for drag polar (details)
  84. ran drag polar cases (details)
  85. refactored reference speed in transition model (details)
  86. typo fix (details)
  87. rename working directory (details)
  88. refator characteristic speed for PTC (details)
  89. refator characteristic speed for PTC (details)
  90. update pyrite files (details)
  91. refactor 3D IBL closure API to allow for alternative closure models (details)
  92. save 3D closure relations in a separate header (details)
  93. add neural net closure for 3D IBL (details)
  94. run drag polar on finer grids (details)
  95. switch to run on acdl desktop (details)
  96. fix memory leak warning caused by accessing index -1 (details)
  97. minor update (details)
  98. mucking with PETSc for IBL3 (details)
  99. instantiate MKL_PARDISO solver for matrix block 3x3 (details)
  100. tried MKL_PARDISO solver for IBL3 (details)
  101. minor typo fix (details)
  102. fix formatting (details)
  103. add mitLG BC for IBL3 (details)
  104. minor refactoring (details)
  105. switch to the same input parameters as DBL3 (details)
  106. add LF flux to IBL3 (details)
  107. add Dirichlet mitLG inflow BC to a Blasius flow test case (details)
  108. intermediate commit: added 4-equation neural net closure model.  working on 4-equation IBL3D pde class (details)
  109. remove extraneous tamplate keyword (details)
  110. added soruce terms for the lateral curvature equation of IBL3 (details)
  111. removed re-definition (details)
  112. refactored and added PDE and BC for four-equation 3D IBL class.  added a starting point for torpedo case (details)
  113. separated classes of 3D IBL and IBLFourEqn (details)
  114. generalized templates in IBL3-related integrands (details)
  115. added instantiations for four-equation 3D IBL (details)
  116. fixed instantiation typo (details)
  117. turned on verbose linesearch printout (details)
  118. added a visous-only Blasius case for 4-equation 3D IBL.  added a missing include (details)
  119. refactored reference coordinate gradient for speedup.  add/edit comments for clarity (details)
  120. typo fix: gradqe*E_qesq.  minor edits (details)
  121. turned off printout in unit testing (details)
  122. cosmetic changes (details)
  123. refactored neural net closure models: H22, Hd1, Hc2 (details)
  124. added output for four-equation 3D IBL (details)
  125. fixed auxi equation (wall transpiration): previous mass flux missed crossflow component (details)
  126. previous fix to wall transpiration (auxi equation) didn't fix.  it is now fixed.  added unit test (details)
  127. minor tweak (details)
  128. minor cleanup (details)
  129. add velocity gradient output to IBL (details)
  130. add instantiation (details)
  131. added IBL3 closure relations in the form of analytic formulae (details)
  132. refactored neural net closure implementation (details)
  133. added a set of IBL3 closure models based on analytic formulae (details)
  134. restored four-equation neural net closure to its original working state (details)
  135. switch to using 4-equation IBL3 closure based on analytic formulae.  results are similar or slightly better than neural net closure (details)
  136. fixed a typo (which fortunately wasn't used yet).  added exception throws (details)
  137. fixed include filename (details)
  138. removed redundant exceptions (details)
  139. refactored four-eqn IBL3 curve-fit closure (details)
  140. conformed to vera check: should not use "using namespace" in header files (details)
  141. fixed unit tests for after API refactoring (details)
  142. fixed unit tests for the  previous API refactoring (details)
  143. fixed ping tests: finite difference step size (details)
  144. added ping test for four-eqn IBL3 (details)
  145. added Blasius flow test case for four-eqn IBL3D. some cleanup (details)
  146. added include (details)
  147. added includes (details)
  148. checked four-eqn IBL3D on Blasius case and stagnation radial flow (details)
  149. added a stagnation flow toy case.  mucked with IBL3 for torpedo case (details)
  150. fixed toy case to keep up with API refactoring (details)
  151. added missing include related to MPI (details)
  152. suppresed file dump in system test (details)
  153. added assertions (details)
  154. add some derivates to the IBL3 output (details)
  155. relaxed pyrite tolerance (details)
  156. relaxed pyrite tolerance for intel compiler.  speed up ping test to avoid time-out (details)
  157. renamed output (details)
  158. added and refactored IBL3 output (details)
  159. trying to fix gnu9 compiler warning (details)
  160. add a new curve fit model based on low-dissipative DBL3 data (details)
  161. allow for uncoupled/viscou-only IBL3 solution by tweaking the panel method (details)
  162. forgot to save changes in the last commit... (details)
  163. relaxed Jenkins timeout threshold for some large ping tests (details)
  164. minor tweak (details)
  165. speed up ping test by turning off some redundant sub-block checks (details)
  166. fix cppcheck error (details)
  167. speed up ping test by turning off tests on some sub-blocks of the jacobian (details)
  168. Don't use python2 with jenkins (details)

Started by upstream project SANS_Weekly build number 264
originally caused by:

Revision: 250c37c7787ab1a0682157d1721701a7093dd34c
Repository: ssh://jenkins@acdl.mit.edu/home/svnroot/SANSproto/
  • acdl/apprentice
  • acdl/shunz/3d_ibl
  • acdl/master
Static Analysis: No warnings
  • Static analysis results from: GNU C Compiler (gcc) (0),  Vera++ (0),  Clang (0),  Intel Compiler (C, Fortran) (0)
  • No issues for 6 builds, i.e. since build: Success #259
  • Reference build: Success release_gnu5,reynolds #263
  • Quality gate: Success
0 errors, 0 warnings
Test Result (no failures)