223031012
/
DAMASK_EICMD
mirror of https://github.com/achalhp/DAMASK_EICMD.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
DAMASK_EICMD/src/mesh_grid.f90

1918 lines
82 KiB
Fortran
Raw Normal View History

introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
!> @author Franz Roters, Max-Planck-Institut für Eisenforschung GmbH
!> @author Philip Eisenlohr, Max-Planck-Institut für Eisenforschung GmbH
!> @author Christoph Koords, Max-Planck-Institut für Eisenforschung GmbH
!> @author Martin Diehl, Max-Planck-Institut für Eisenforschung GmbH
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
!> @brief Sets up the mesh for the solvers MSC.Marc, Abaqus and the spectral solver
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
makefile modified not to link against wrong libraries when using PETSc, reading in of values in phenopowerlaw improved, HDF5 in mesh
2014-04-29 23:20:59 +05:30
module mesh
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
use, intrinsic :: iso_c_binding
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
use prec, only: pReal, pInt
clean and initialize element
2019-01-24 19:49:17 +05:30
use mesh_base
moved calculation and output of geometry related data to mesh.f90, moved output of command line related information to DAMASK_spectral_interface.f90. improved new solver structure (use with make SOLVER=NEW)
2012-08-06 18:13:05 +05:30
added function mesh_FEtoCPelement(FEid)
2007-03-21 21:48:33 +05:30
implicit none
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
private
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
integer(pInt), public, protected :: &
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
mesh_NcpElems, & !< total number of CP elements in local mesh
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
mesh_Nnodes, & !< total number of nodes in mesh
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
mesh_Ncellnodes, & !< total number of cell nodes in mesh (including duplicates)
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
mesh_Ncells, & !< total number of cells in mesh
introducing better names allows further simplifications as we do not store max and per elem values any more for number of integration points and number of cell nodes
2018-09-23 21:07:57 +05:30
mesh_NcellnodesPerElem, & !< number of cell nodes per element
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
mesh_maxNipNeighbors, & !< max number of IP neighbors in any CP element
introducing better names allows further simplifications as we do not store max and per elem values any more for number of integration points and number of cell nodes
2018-09-23 21:07:57 +05:30
mesh_maxNsharedElems !< max number of CP elements sharing a node
!!!! BEGIN DEPRECATED !!!!!
integer(pInt), public, protected :: &
Nelems /= NcpElems only in case of Abaqus/MSC.Marc
2018-09-23 20:35:01 +05:30
mesh_maxNcellnodes !< max number of cell nodes in any CP element
introducing better names allows further simplifications as we do not store max and per elem values any more for number of integration points and number of cell nodes
2018-09-23 21:07:57 +05:30
!!!! BEGIN DEPRECATED !!!!!
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
integer(pInt), dimension(:), allocatable, private :: &
microGlobal
more descriptive names
2018-09-23 22:20:54 +05:30
integer(pInt), dimension(:), allocatable, public, protected :: &
consistent (and better understandable) names
2018-10-04 09:33:48 +05:30
mesh_homogenizationAt, & !< homogenization ID of each element
mesh_microstructureAt !< microstructure ID of each element
more descriptive names
2018-09-23 22:20:54 +05:30
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
integer(pInt), dimension(:,:), allocatable, public, protected :: &
unused variables
2019-02-02 16:53:53 +05:30
mesh_element !< entryCount and list of elements containing node
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
integer(pInt), dimension(:,:,:,:), allocatable, public, protected :: &
third entry in mesh_ipNeighborhood stores the neighbor_index that points from each neighbor back to the central ip; needed in nonlocal model
2012-10-24 19:33:02 +05:30
mesh_ipNeighborhood !< 6 or less neighboring IPs as [element_num, IP_index, neighbor_index that points to me]
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
"unitlength" parameter, which determines the physical size of the mesh, now available as a global mesh variable "mesh_unitlength" and written to the .mesh file during init. Hence, it is available to the marc_deformedGeometry.py script via "mesh_init_postprocessing()" and "mesh_get_unitlength()", so no need for setting the scaling of the displacement vectors explicitly through an option; now displacements and nodal positions are always consistent
2013-05-07 18:36:29 +05:30
real(pReal), public, protected :: &
mesh_unitlength !< physical length of one unit in mesh
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
real(pReal), dimension(:,:), allocatable, public :: &
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_node, & !< node x,y,z coordinates (after deformation! ONLY FOR MARC!!!)
mesh_cellnode !< cell node x,y,z coordinates (after deformation! ONLY FOR MARC!!!)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
real(pReal), dimension(:,:), allocatable, public, protected :: &
mesh_ipVolume, & !< volume associated with IP (initially!)
mesh_node0 !< node x,y,z coordinates (initially!)
real(pReal), dimension(:,:,:), allocatable, public, protected :: &
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
mesh_ipArea !< area of interface to neighboring IP (initially!)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
real(pReal), dimension(:,:,:), allocatable, public :: &
mesh_ipCoordinates !< IP x,y,z coordinates (after deformation!)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
real(pReal),dimension(:,:,:,:), allocatable, public, protected :: &
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
mesh_ipAreaNormal !< area normal of interface to neighboring IP (initially!)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
all variables/functions were not used
2019-01-31 16:53:23 +05:30
logical, dimension(3), public, parameter :: mesh_periodicSurface = .true. !< flag indicating periodic outer surfaces (used for fluxes)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
grouping solver specific variables for better readability
2018-09-23 18:57:51 +05:30
integer(pInt), dimension(:,:), allocatable, private :: &
introduced preprocessor statements in mesh to hide unneeded variables as a preparation for HDF5
2013-12-28 01:33:28 +05:30
mesh_cellnodeParent !< cellnode's parent element ID, cellnode's intra-element ID
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt),dimension(:,:,:), allocatable, private :: &
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_cell !< cell connectivity for each element,ip/cell
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(:,:,:), allocatable, private :: &
FE_nodesAtIP, & !< map IP index to node indices in a specific type of element
FE_ipNeighbor, & !< +x,-x,+y,-y,+z,-z list of intra-element IPs and(negative) neighbor faces per own IP in a specific type of element
FE_cell, & !< list of intra-element cell node IDs that constitute the cells in a specific type of element geometry
FE_cellface !< list of intra-cell cell node IDs that constitute the cell faces of a specific type of cell
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
real(pReal), dimension(:,:,:), allocatable, private :: &
FE_cellnodeParentnodeWeights !< list of node weights for the generation of cell nodes
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
! These definitions should actually reside in the FE-solver specific part (different for MARC/ABAQUS)
started envIP building
2007-03-22 20:18:58 +05:30
! Hence, I suggest to prefix with "FE_"
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
integer(pInt), parameter, public :: &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_Nelemtypes = 13_pInt, &
FE_Ngeomtypes = 10_pInt, &
FE_Ncelltypes = 4_pInt, &
FE_maxNnodes = 20_pInt, &
FE_maxNips = 27_pInt, &
FE_maxNipNeighbors = 6_pInt, &
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
FE_maxmaxNnodesAtIP = 8_pInt, & !< max number of (equivalent) nodes attached to an IP
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_maxNmatchingNodesPerFace = 4_pInt, &
FE_maxNfaces = 6_pInt, &
FE_maxNcellnodes = 64_pInt, &
FE_maxNcellnodesPerCell = 8_pInt, &
FE_maxNcellfaces = 6_pInt, &
FE_maxNcellnodesPerCellface = 4_pInt
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
integer(pInt), dimension(FE_Nelemtypes), parameter, public :: FE_geomtype = & !< geometry type of particular element type
int([ &
1, & ! element 6 (2D 3node 1ip)
2, & ! element 125 (2D 6node 3ip)
3, & ! element 11 (2D 4node 4ip)
4, & ! element 27 (2D 8node 9ip)
- added element type 54 for macro (2d 8node 4ip reduced integration) - automatically raising error for unknown element type inside mapping function
2013-04-10 15:08:40 +05:30
3, & ! element 54 (2D 8node 4ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
5, & ! element 134 (3D 4node 1ip)
6, & ! element 157 (3D 5node 4ip)
6, & ! element 127 (3D 10node 4ip)
7, & ! element 136 (3D 6node 6ip)
8, & ! element 117 (3D 8node 1ip)
9, & ! element 7 (3D 8node 8ip)
9, & ! element 57 (3D 20node 8ip)
10 & ! element 21 (3D 20node 27ip)
polishing: removed variable names like 'unit' and 'data' that are keywords of fortran and ensured that integer and real precision matches independent of machine standard. removed cut_off parameter for damask_spectral removed outpot of derived divergence measures and added RMS output in brackets added comments and options to the makefile
2012-02-16 00:28:38 +05:30
],pInt)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(FE_Ngeomtypes), parameter, public :: FE_celltype = & !< cell type that is used by each geometry type
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
int([ &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, & ! element 6 (2D 3node 1ip)
2, & ! element 125 (2D 6node 3ip)
2, & ! element 11 (2D 4node 4ip)
2, & ! element 27 (2D 8node 9ip)
3, & ! element 134 (3D 4node 1ip)
4, & ! element 127 (3D 10node 4ip)
4, & ! element 136 (3D 6node 6ip)
4, & ! element 117 (3D 8node 1ip)
4, & ! element 7 (3D 8node 8ip)
4 & ! element 21 (3D 20node 27ip)
polishing: removed variable names like 'unit' and 'data' that are keywords of fortran and ensured that integer and real precision matches independent of machine standard. removed cut_off parameter for damask_spectral removed outpot of derived divergence measures and added RMS output in brackets added comments and options to the makefile
2012-02-16 00:28:38 +05:30
],pInt)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(FE_Ngeomtypes), parameter, public :: FE_dimension = & !< dimension of geometry type
introduced dedicated calculation of interface normals in IP neighborhood. (2D elements did not work so far..!) fixed definitions of some (2D) interface planes to have all normals pointing outward. removed "int, private :: i" defined for spectral compilation.
2013-02-05 18:57:37 +05:30
int([ &
2, & ! element 6 (2D 3node 1ip)
2, & ! element 125 (2D 6node 3ip)
2, & ! element 11 (2D 4node 4ip)
2, & ! element 27 (2D 8node 9ip)
3, & ! element 134 (3D 4node 1ip)
3, & ! element 127 (3D 10node 4ip)
3, & ! element 136 (3D 6node 6ip)
3, & ! element 117 (3D 8node 1ip)
3, & ! element 7 (3D 8node 8ip)
3 & ! element 21 (3D 20node 27ip)
],pInt)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(FE_Nelemtypes), parameter, public :: FE_Nnodes = & !< number of nodes that constitute a specific type of element
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
int([ &
3, & ! element 6 (2D 3node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
6, & ! element 125 (2D 6node 3ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
4, & ! element 11 (2D 4node 4ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
8, & ! element 27 (2D 8node 9ip)
8, & ! element 54 (2D 8node 4ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
4, & ! element 134 (3D 4node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
5, & ! element 157 (3D 5node 4ip)
10, & ! element 127 (3D 10node 4ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
6, & ! element 136 (3D 6node 6ip)
8, & ! element 117 (3D 8node 1ip)
8, & ! element 7 (3D 8node 8ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
20, & ! element 57 (3D 20node 8ip)
20 & ! element 21 (3D 20node 27ip)
polishing: removed variable names like 'unit' and 'data' that are keywords of fortran and ensured that integer and real precision matches independent of machine standard. removed cut_off parameter for damask_spectral removed outpot of derived divergence measures and added RMS output in brackets added comments and options to the makefile
2012-02-16 00:28:38 +05:30
],pInt)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(FE_Ngeomtypes), parameter, public :: FE_Nfaces = & !< number of faces of a specific type of element geometry
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
int([ &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
3, & ! element 6 (2D 3node 1ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
3, & ! element 125 (2D 6node 3ip)
4, & ! element 11 (2D 4node 4ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
4, & ! element 27 (2D 8node 9ip)
4, & ! element 134 (3D 4node 1ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
4, & ! element 127 (3D 10node 4ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
5, & ! element 136 (3D 6node 6ip)
6, & ! element 117 (3D 8node 1ip)
6, & ! element 7 (3D 8node 8ip)
6 & ! element 21 (3D 20node 27ip)
polishing: removed variable names like 'unit' and 'data' that are keywords of fortran and ensured that integer and real precision matches independent of machine standard. removed cut_off parameter for damask_spectral removed outpot of derived divergence measures and added RMS output in brackets added comments and options to the makefile
2012-02-16 00:28:38 +05:30
],pInt)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(FE_Ngeomtypes), parameter, private :: FE_NmatchingNodes = & !< number of nodes that are needed for face matching in a specific type of element geometry
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
int([ &
3, & ! element 6 (2D 3node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
3, & ! element 125 (2D 6node 3ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
4, & ! element 11 (2D 4node 4ip)
4, & ! element 27 (2D 8node 9ip)
4, & ! element 134 (3D 4node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
4, & ! element 127 (3D 10node 4ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
6, & ! element 136 (3D 6node 6ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
8, & ! element 117 (3D 8node 1ip)
8, & ! element 7 (3D 8node 8ip)
8 & ! element 21 (3D 20node 27ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
],pInt)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(FE_Ngeomtypes), parameter, private :: FE_Ncellnodes = & !< number of cell nodes in a specific geometry type
int([ &
3, & ! element 6 (2D 3node 1ip)
7, & ! element 125 (2D 6node 3ip)
9, & ! element 11 (2D 4node 4ip)
fixed bug in cell generation: "FE_Ncellnodes" was not correct for elem type 17 and 136 too much memory allocated for mesh_cell
2013-04-18 23:26:14 +05:30
16, & ! element 27 (2D 8node 9ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
4, & ! element 134 (3D 4node 1ip)
15, & ! element 127 (3D 10node 4ip)
fixed bug in cell generation: "FE_Ncellnodes" was not correct for elem type 17 and 136 too much memory allocated for mesh_cell
2013-04-18 23:26:14 +05:30
21, & ! element 136 (3D 6node 6ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
8, & ! element 117 (3D 8node 1ip)
27, & ! element 7 (3D 8node 8ip)
64 & ! element 21 (3D 20node 27ip)
],pInt)
integer(pInt), dimension(FE_Ncelltypes), parameter, private :: FE_NcellnodesPerCell = & !< number of cell nodes in a specific cell type
int([ &
3, & ! (2D 3node)
4, & ! (2D 4node)
4, & ! (3D 4node)
8 & ! (3D 8node)
],pInt)
integer(pInt), dimension(FE_Ncelltypes), parameter, private :: FE_NcellnodesPerCellface = & !< number of cell nodes per cell face in a specific cell type
int([&
2, & ! (2D 3node)
2, & ! (2D 4node)
3, & ! (3D 4node)
4 & ! (3D 8node)
],pInt)
integer(pInt), dimension(FE_Ngeomtypes), parameter, public :: FE_Nips = & !< number of IPs in a specific type of element
int([ &
1, & ! element 6 (2D 3node 1ip)
3, & ! element 125 (2D 6node 3ip)
4, & ! element 11 (2D 4node 4ip)
9, & ! element 27 (2D 8node 9ip)
1, & ! element 134 (3D 4node 1ip)
4, & ! element 127 (3D 10node 4ip)
6, & ! element 136 (3D 6node 6ip)
1, & ! element 117 (3D 8node 1ip)
8, & ! element 7 (3D 8node 8ip)
27 & ! element 21 (3D 20node 27ip)
],pInt)
integer(pInt), dimension(FE_Ncelltypes), parameter, public :: FE_NipNeighbors = & !< number of ip neighbors / cell faces in a specific cell type
int([&
3, & ! (2D 3node)
4, & ! (2D 4node)
4, & ! (3D 4node)
6 & ! (3D 8node)
],pInt)
integer(pInt), dimension(FE_Ngeomtypes), parameter, private :: FE_maxNnodesAtIP = & !< maximum number of parent nodes that belong to an IP for a specific type of element
int([ &
3, & ! element 6 (2D 3node 1ip)
1, & ! element 125 (2D 6node 3ip)
1, & ! element 11 (2D 4node 4ip)
2, & ! element 27 (2D 8node 9ip)
4, & ! element 134 (3D 4node 1ip)
1, & ! element 127 (3D 10node 4ip)
1, & ! element 136 (3D 6node 6ip)
8, & ! element 117 (3D 8node 1ip)
1, & ! element 7 (3D 8node 8ip)
4 & ! element 21 (3D 20node 27ip)
],pInt)
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
grouping solver specific variables for better readability
2018-09-23 18:57:51 +05:30
integer(pInt), dimension(3), public, protected :: &
grid !< (global) grid
integer(pInt), public, protected :: &
mesh_NcpElemsGlobal, & !< total number of CP elements in global mesh
grid3, & !< (local) grid in 3rd direction
grid3Offset !< (local) grid offset in 3rd direction
real(pReal), dimension(3), public, protected :: &
geomSize
real(pReal), public, protected :: &
size3, & !< (local) size in 3rd direction
size3offset !< (local) size offset in 3rd direction
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
public :: &
mesh_init, &
all variables/functions were not used
2019-01-31 16:53:23 +05:30
mesh_cellCenterCoordinates
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
private :: &
consistent order of solver-specific functionality
2018-09-23 21:27:48 +05:30
mesh_build_cellconnectivity, &
mesh_build_ipAreas, &
mesh_build_FEdata, &
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
mesh_spectral_getHomogenization, &
mesh_spectral_build_nodes, &
mesh_spectral_build_elements, &
all variables/functions were not used
2019-01-31 16:53:23 +05:30
mesh_spectral_build_ipNeighborhood, &
mesh_build_cellnodes, &
mesh_build_ipVolumes, &
mesh_build_ipCoordinates
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
clean and initialize element
2019-01-24 19:49:17 +05:30
type, public, extends(tMesh) :: tMesh_grid
initialize element
2019-01-24 20:47:20 +05:30
integer(pInt), dimension(3), public :: &
grid !< (global) grid
integer(pInt), public :: &
mesh_NcpElemsGlobal, & !< total number of CP elements in global mesh
grid3, & !< (local) grid in 3rd direction
grid3Offset !< (local) grid offset in 3rd direction
real(pReal), dimension(3), public :: &
geomSize
real(pReal), public :: &
size3, & !< (local) size in 3rd direction
size3offset
clean and initialize element
2019-01-24 19:49:17 +05:30
contains
initialize mesh and element
2019-02-01 16:54:23 +05:30
procedure, pass(self) :: tMesh_grid_init
generic, public :: init => tMesh_grid_init
clean and initialize element
2019-01-24 19:49:17 +05:30
end type tMesh_grid
initialize element
2019-01-24 20:47:20 +05:30
type(tMesh_grid), public, protected :: theMesh
clean and initialize element
2019-01-24 19:49:17 +05:30
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
contains
added function mesh_FEtoCPelement(FEid)
2007-03-21 21:48:33 +05:30
initialize mesh and element
2019-02-01 16:54:23 +05:30
subroutine tMesh_grid_init(self,nodes)
clean and initialize element
2019-01-24 19:49:17 +05:30
implicit none
class(tMesh_grid) :: self
initialize mesh and element
2019-02-01 16:54:23 +05:30
real(pReal), dimension(:,:), intent(in) :: nodes
clean and initialize element
2019-01-24 19:49:17 +05:30
initialize mesh and element
2019-02-01 16:54:23 +05:30
call self%tMesh%init('grid',10_pInt,nodes)
clean and initialize element
2019-01-24 19:49:17 +05:30
end subroutine tMesh_grid_init
major debugging and streamlining of the calculation of ipVolumes and ipAreas
2009-01-20 00:12:31 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief initializes the mesh by calling all necessary private routines the mesh module
!! Order and routines strongly depend on type of solver
!--------------------------------------------------------------------------------------------------
introduced dedicated calculation of interface normals in IP neighborhood. (2D elements did not work so far..!) fixed definitions of some (2D) interface planes to have all normals pointing outward. removed "int, private :: i" defined for spectral compilation.
2013-02-05 18:57:37 +05:30
subroutine mesh_init(ip,el)
compiler_options and compiler_version supported by new Intel compiler
2018-02-02 17:06:09 +05:30
#if defined(__GFORTRAN__) || __INTEL_COMPILER >= 1800
suppress warning when compiling with gfortran
2017-10-05 20:05:34 +05:30
use, intrinsic :: iso_fortran_env, only: &
compiler_version, &
compiler_options
employing use statements (almost no external functions anymore) adopted function calls to newer PETSc versions
2018-05-17 15:34:21 +05:30
#endif
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
employing use statements (almost no external functions anymore) adopted function calls to newer PETSc versions
2018-05-17 15:34:21 +05:30
#include <petsc/finclude/petscsys.h>
use PETScsys
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
no need to inlcude twice
2017-02-05 04:23:22 +05:30
use DAMASK_interface
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
use IO, only: &
use statement for marc was wrong (preprocessor)
2015-04-11 10:39:15 +05:30
IO_open_file, &
error check for mpi parallelization
2016-08-16 17:00:11 +05:30
IO_error, &
use statement for marc was wrong (preprocessor)
2015-04-11 10:39:15 +05:30
IO_timeStamp, &
IO_error, &
IO_write_jobFile
introduced error when element/IP/component(grain) selected for debugging does not exist
2013-10-16 18:08:00 +05:30
use debug, only: &
debug_e, &
added some debug statements
2014-12-19 00:11:02 +05:30
debug_i, &
debug_level, &
debug_mesh, &
debug_levelBasic
further improvements on implementing the plain mode
2013-04-16 22:37:27 +05:30
use numerics, only: &
all variables/functions were not used
2019-01-31 16:53:23 +05:30
numerics_unitlength
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
use FEsolving, only: &
syntax errors in declaration
2018-09-23 22:34:17 +05:30
FEsolving_execElem, &
FEsolving_execIP
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
added function calls in init()
2007-04-10 16:52:53 +05:30
implicit none
employing use statements (almost no external functions anymore) adopted function calls to newer PETSc versions
2018-05-17 15:34:21 +05:30
include 'fftw3-mpi.f03'
error check for mpi parallelization
2016-08-16 17:00:11 +05:30
integer(C_INTPTR_T) :: devNull, local_K, local_K_offset
integer :: ierr, worldsize
calcMode not needed for spectral and FEM
2018-09-24 01:01:30 +05:30
integer(pInt), intent(in), optional :: el, ip
improved reading in of values, now only warnings in case of problematic entries in material.config divergence calculation sqrt scaling optionally introduced for basic scheme spectral solver
2013-02-08 21:25:53 +05:30
integer(pInt) :: j
added some debug statements
2014-12-19 00:11:02 +05:30
logical :: myDebug
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
removed old regridding related stuff
2016-06-29 20:29:22 +05:30
write(6,'(/,a)') ' <<<+- mesh init -+>>>'
write(6,'(a15,a)') ' Current time: ',IO_timeStamp()
made code standard conform to Fortran 2008 (ignoring warning concerning comments beyond character 132). Basically, changing "x" format specifier to "Nx" ("1x") plus removing $ format specifier added compiler switches for gfortran and ifort to check for standard conformity old gnu compilers <4.4 are not longer supported because they don't provide the c binding for fftw
2012-02-01 00:48:55 +05:30
#include "compilation_info.f90"
removed unnecsessary omp statements, removed one transpose in constitutitve
2013-01-08 16:39:20 +05:30
initialize mesh and element
2019-02-01 16:54:23 +05:30
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
call mesh_build_FEdata ! get properties of the different types of elements
"unitlength" parameter, which determines the physical size of the mesh, now available as a global mesh variable "mesh_unitlength" and written to the .mesh file during init. Hence, it is available to the marc_deformedGeometry.py script via "mesh_init_postprocessing()" and "mesh_get_unitlength()", so no need for setting the scaling of the displacement vectors explicitly through an option; now displacements and nodal positions are always consistent
2013-05-07 18:36:29 +05:30
mesh_unitlength = numerics_unitlength ! set physical extent of a length unit in mesh
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
added some debug statements
2014-12-19 00:11:02 +05:30
myDebug = (iand(debug_level(debug_mesh),debug_levelBasic) /= 0_pInt)
"unitlength" parameter, which determines the physical size of the mesh, now available as a global mesh variable "mesh_unitlength" and written to the .mesh file during init. Hence, it is available to the marc_deformedGeometry.py script via "mesh_init_postprocessing()" and "mesh_get_unitlength()", so no need for setting the scaling of the displacement vectors explicitly through an option; now displacements and nodal positions are always consistent
2013-05-07 18:36:29 +05:30
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
call fftw_mpi_init()
only parse geom file once
2019-02-02 13:48:01 +05:30
call mesh_spectral_read_grid()
consistent use of same name for communicator will not make any difference, but clearer to understand http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Sys/PETSC_COMM_WORLD.html
2016-10-24 14:03:01 +05:30
call MPI_comm_size(PETSC_COMM_WORLD, worldsize, ierr)
error check for mpi parallelization
2016-08-16 17:00:11 +05:30
if(ierr /=0_pInt) call IO_error(894_pInt, ext_msg='MPI_comm_size')
if(worldsize>grid(3)) call IO_error(894_pInt, ext_msg='number of processes exceeds grid(3)')
only parse geom file once
2019-02-02 13:48:01 +05:30
polishing and encapsulating of Abaqus-specific local variables
2018-10-10 03:22:54 +05:30
devNull = fftw_mpi_local_size_3d(int(grid(3),C_INTPTR_T), &
int(grid(2),C_INTPTR_T), &
int(grid(1),C_INTPTR_T)/2+1, &
PETSC_COMM_WORLD, &
local_K, & ! domain grid size along z
local_K_offset) ! domain grid offset along z
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
grid3 = int(local_K,pInt)
grid3Offset = int(local_K_offset,pInt)
size3 = geomSize(3)*real(grid3,pReal) /real(grid(3),pReal)
size3Offset = geomSize(3)*real(grid3Offset,pReal)/real(grid(3),pReal)
initialize mesh and element
2019-02-01 16:54:23 +05:30
mesh_NcpElems= product(grid(1:2))*grid3
mesh_NcpElemsGlobal = product(grid)
mesh_Nnodes = product(grid(1:2) + 1_pInt)*(grid3 + 1_pInt)
took some gfortran complaints serious (unused imports, implicit castings)
2015-04-11 17:17:33 +05:30
call mesh_spectral_build_nodes()
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built nodes'; flush(6)
only parse geom file once
2019-02-02 13:48:01 +05:30
call theMesh%init(mesh_node)
initialize mesh and element
2019-02-01 16:54:23 +05:30
! For compatibility
mesh_maxNipNeighbors = theMesh%elem%nIPneighbors
mesh_maxNcellnodes = theMesh%elem%Ncellnodes
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
call theMesh%setNelems(mesh_NcpElems)
initialize mesh and element
2019-02-01 16:54:23 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
call mesh_spectral_build_elements()
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built elements'; flush(6)
only parse geom file once
2019-02-02 13:48:01 +05:30
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
call mesh_build_cellconnectivity
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built cell connectivity'; flush(6)
added python script that produces vtk files with the (node based and ip based) deformed mesh from marc output file
2013-04-24 00:00:56 +05:30
mesh_cellnode = mesh_build_cellnodes(mesh_node,mesh_Ncellnodes)
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built cell nodes'; flush(6)
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
call mesh_build_ipCoordinates
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built IP coordinates'; flush(6)
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
call mesh_build_ipVolumes
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built IP volumes'; flush(6)
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
call mesh_build_ipAreas
added some debug statements
2014-12-19 00:11:02 +05:30
if (myDebug) write(6,'(a)') ' Built IP areas'; flush(6)
some simplifications as discussed with IBF colleagues
2015-11-26 02:25:17 +05:30
removed a bunch of unneded functions
2016-06-27 21:45:56 +05:30
call mesh_spectral_build_ipNeighborhood
reordered (and partly redistributed) error message identifiers, deleted those which are not in use anymore. all constitutive as well numerics now raises an error, if an unknown keyword is found in the respective config file
2012-02-13 23:11:27 +05:30
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
if (myDebug) write(6,'(a)') ' Built IP neighborhood'; flush(6)
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
introduced error when element/IP/component(grain) selected for debugging does not exist
2013-10-16 18:08:00 +05:30
if (debug_e < 1 .or. debug_e > mesh_NcpElems) &
call IO_error(602_pInt,ext_msg='element') ! selected element does not exist
if (debug_i < 1 .or. debug_i > FE_Nips(FE_geomtype(mesh_element(2_pInt,debug_e)))) &
call IO_error(602_pInt,ext_msg='IP') ! selected element does not have requested IP
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
added few comments
2014-01-17 07:08:35 +05:30
FEsolving_execElem = [ 1_pInt,mesh_NcpElems ] ! parallel loop bounds set to comprise all DAMASK elements
leaner syntax with sourced allocation
2018-09-23 20:56:13 +05:30
allocate(FEsolving_execIP(2_pInt,mesh_NcpElems), source=1_pInt) ! parallel loop bounds set to comprise from first IP...
added few comments
2014-01-17 07:08:35 +05:30
forall (j = 1_pInt:mesh_NcpElems) FEsolving_execIP(2,j) = FE_Nips(FE_geomtype(mesh_element(2,j))) ! ...up to own IP count for each element
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
fixed small bug when calculating mesh_Ncells
2013-04-19 18:11:06 +05:30
introducing better names allows further simplifications as we do not store max and per elem values any more for number of integration points and number of cell nodes
2018-09-23 21:07:57 +05:30
!!!! COMPATIBILITY HACK !!!!
! for a homogeneous mesh, all elements have the same number of IPs and and cell nodes.
! hence, xxPerElem instead of maxXX
mesh_NcellnodesPerElem = mesh_maxNcellnodes
more descriptive names
2018-09-23 22:20:54 +05:30
! better name
consistent (and better understandable) names
2018-10-04 09:33:48 +05:30
mesh_homogenizationAt = mesh_element(3,:)
mesh_microstructureAt = mesh_element(4,:)
introducing better names allows further simplifications as we do not store max and per elem values any more for number of integration points and number of cell nodes
2018-09-23 21:07:57 +05:30
!!!!!!!!!!!!!!!!!!!!!!!!
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine mesh_init
major debugging and streamlining of the calculation of ipVolumes and ipAreas
2009-01-20 00:12:31 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!> @brief Split CP elements into cells.
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
!> @details Build a mapping between cells and the corresponding cell nodes ('mesh_cell').
!> Cell nodes that are also matching nodes are unique in the list of cell nodes,
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
!> all others (currently) might be stored more than once.
!> Also allocates the 'mesh_node' array.
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
subroutine mesh_build_cellconnectivity
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt), dimension(:), allocatable :: &
matchingNode2cellnode
integer(pInt), dimension(:,:), allocatable :: &
cellnodeParent
integer(pInt), dimension(mesh_maxNcellnodes) :: &
localCellnode2globalCellnode
introduced error when element/IP/component(grain) selected for debugging does not exist
2013-10-16 18:08:00 +05:30
integer(pInt) :: &
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
e,t,g,c,n,i, &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
matchingNodeID, &
localCellnodeID
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
allocate(mesh_cell(FE_maxNcellnodesPerCell,theMesh%elem%nIPs,mesh_NcpElems), source=0_pInt)
added some code parts for HDF5 in Fortran
2013-12-30 15:36:01 +05:30
allocate(matchingNode2cellnode(mesh_Nnodes), source=0_pInt)
allocate(cellnodeParent(2_pInt,mesh_maxNcellnodes*mesh_NcpElems), source=0_pInt)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced error when element/IP/component(grain) selected for debugging does not exist
2013-10-16 18:08:00 +05:30
!--------------------------------------------------------------------------------------------------
! Count cell nodes (including duplicates) and generate cell connectivity list
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_Ncellnodes = 0_pInt
mesh_Ncells = 0_pInt
do e = 1_pInt,mesh_NcpElems ! loop over cpElems
t = mesh_element(2_pInt,e) ! get element type
g = FE_geomtype(t) ! get geometry type
c = FE_celltype(g) ! get cell type
localCellnode2globalCellnode = 0_pInt
mesh_Ncells = mesh_Ncells + FE_Nips(g)
do i = 1_pInt,FE_Nips(g) ! loop over ips=cells in this element
do n = 1_pInt,FE_NcellnodesPerCell(c) ! loop over cell nodes in this cell
localCellnodeID = FE_cell(n,i,g)
if (localCellnodeID <= FE_NmatchingNodes(g)) then ! this cell node is a matching node
matchingNodeID = mesh_element(4_pInt+localCellnodeID,e)
if (matchingNode2cellnode(matchingNodeID) == 0_pInt) then ! if this matching node does not yet exist in the glbal cell node list ...
mesh_Ncellnodes = mesh_Ncellnodes + 1_pInt ! ... count it as cell node ...
matchingNode2cellnode(matchingNodeID) = mesh_Ncellnodes ! ... and remember its global ID
cellnodeParent(1_pInt,mesh_Ncellnodes) = e ! ... and where it belongs to
cellnodeParent(2_pInt,mesh_Ncellnodes) = localCellnodeID
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
endif
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_cell(n,i,e) = matchingNode2cellnode(matchingNodeID)
else ! this cell node is no matching node
if (localCellnode2globalCellnode(localCellnodeID) == 0_pInt) then ! if this local cell node does not yet exist in the global cell node list ...
mesh_Ncellnodes = mesh_Ncellnodes + 1_pInt ! ... count it as cell node ...
localCellnode2globalCellnode(localCellnodeID) = mesh_Ncellnodes ! ... and remember its global ID ...
cellnodeParent(1_pInt,mesh_Ncellnodes) = e ! ... and it belongs to
cellnodeParent(2_pInt,mesh_Ncellnodes) = localCellnodeID
endif
mesh_cell(n,i,e) = localCellnode2globalCellnode(localCellnodeID)
endif
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
enddo
enddo
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
enddo
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
allocate(mesh_cellnodeParent(2_pInt,mesh_Ncellnodes))
allocate(mesh_cellnode(3_pInt,mesh_Ncellnodes))
forall(n = 1_pInt:mesh_Ncellnodes)
mesh_cellnodeParent(1,n) = cellnodeParent(1,n)
mesh_cellnodeParent(2,n) = cellnodeParent(2,n)
endforall
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
end subroutine mesh_build_cellconnectivity
!--------------------------------------------------------------------------------------------------
!> @brief Calculate position of cellnodes from the given position of nodes
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
!> Build list of cellnodes' coordinates.
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
!> Cellnode coordinates are calculated from a weighted sum of node coordinates.
!--------------------------------------------------------------------------------------------------
added python script that produces vtk files with the (node based and ip based) deformed mesh from marc output file
2013-04-24 00:00:56 +05:30
function mesh_build_cellnodes(nodes,Ncellnodes)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
implicit none
introduced error when element/IP/component(grain) selected for debugging does not exist
2013-10-16 18:08:00 +05:30
integer(pInt), intent(in) :: Ncellnodes !< requested number of cellnodes
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
real(pReal), dimension(3,mesh_Nnodes), intent(in) :: nodes
added python script that produces vtk files with the (node based and ip based) deformed mesh from marc output file
2013-04-24 00:00:56 +05:30
real(pReal), dimension(3,Ncellnodes) :: mesh_build_cellnodes
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
introduced error when element/IP/component(grain) selected for debugging does not exist
2013-10-16 18:08:00 +05:30
integer(pInt) :: &
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
e,t,n,m, &
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
localCellnodeID
real(pReal), dimension(3) :: &
myCoords
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_build_cellnodes = 0.0_pReal
fixed bug in cell generation: "FE_Ncellnodes" was not correct for elem type 17 and 136 too much memory allocated for mesh_cell
2013-04-18 23:26:14 +05:30
!$OMP PARALLEL DO PRIVATE(e,localCellnodeID,t,myCoords)
added python script that produces vtk files with the (node based and ip based) deformed mesh from marc output file
2013-04-24 00:00:56 +05:30
do n = 1_pInt,Ncellnodes ! loop over cell nodes
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
e = mesh_cellnodeParent(1,n)
localCellnodeID = mesh_cellnodeParent(2,n)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
t = mesh_element(2,e) ! get element type
myCoords = 0.0_pReal
do m = 1_pInt,FE_Nnodes(t)
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
myCoords = myCoords + nodes(1:3,mesh_element(4_pInt+m,e)) &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
* FE_cellnodeParentnodeWeights(m,localCellnodeID,t)
enddo
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_build_cellnodes(1:3,n) = myCoords / sum(FE_cellnodeParentnodeWeights(:,localCellnodeID,t))
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
enddo
fixed bug in cell generation: "FE_Ncellnodes" was not correct for elem type 17 and 136 too much memory allocated for mesh_cell
2013-04-18 23:26:14 +05:30
!$OMP END PARALLEL DO
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
end function mesh_build_cellnodes
Now able to have real periodicity for fluxes (fluxes leaving the model on one side enter on the other side). To enable this feature one has to add the following somewhere in the marc input file: $mpie periodic x y z for having periodicity in all directions $mpie periodic z x for having periodicity in x and z direction etc. Note that this only works for regular meshes!!!
2011-02-16 21:53:08 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief Calculates IP volume. Allocates global array 'mesh_ipVolume'
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!> @details The IP volume is calculated differently depending on the cell type.
!> 2D cells assume an element depth of one in order to calculate the volume.
!> For the hexahedral cell we subdivide the cell into subvolumes of pyramidal
!> shape with a cell face as basis and the central ip at the tip. This subvolume is
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
!> calculated as an average of four tetrahedals with three corners on the cell face
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!> and one corner at the central ip.
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
subroutine mesh_build_ipVolumes
allowed arbitrary grid (formerly know as resolution), i.e. any integer>1 in each dimension is possible. removed square root correction for divergence (doesn't make sense) fixed FFT debug of spectral solver
2013-04-10 15:49:16 +05:30
use math, only: &
math_volTetrahedron, &
math_areaTriangle
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt) :: e,t,g,c,i,m,f,n
real(pReal), dimension(FE_maxNcellnodesPerCellface,FE_maxNcellfaces) :: subvolume
Now able to have real periodicity for fluxes (fluxes leaving the model on one side enter on the other side). To enable this feature one has to add the following somewhere in the marc input file: $mpie periodic x y z for having periodicity in all directions $mpie periodic z x for having periodicity in x and z direction etc. Note that this only works for regular meshes!!!
2011-02-16 21:53:08 +05:30
initialize mesh and element
2019-02-01 16:54:23 +05:30
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
allocate(mesh_ipVolume(theMesh%elem%nIPs,mesh_NcpElems),source=0.0_pReal)
initialize mesh and element
2019-02-01 16:54:23 +05:30
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!$OMP PARALLEL DO PRIVATE(t,g,c,m,subvolume)
do e = 1_pInt,mesh_NcpElems ! loop over cpElems
t = mesh_element(2_pInt,e) ! get element type
g = FE_geomtype(t) ! get geometry type
c = FE_celltype(g) ! get cell type
select case (c)
case (1_pInt) ! 2D 3node
forall (i = 1_pInt:FE_Nips(g)) & ! loop over ips=cells in this element
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_ipVolume(i,e) = math_areaTriangle(mesh_cellnode(1:3,mesh_cell(1,i,e)), &
mesh_cellnode(1:3,mesh_cell(2,i,e)), &
mesh_cellnode(1:3,mesh_cell(3,i,e)))
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
case (2_pInt) ! 2D 4node
forall (i = 1_pInt:FE_Nips(g)) & ! loop over ips=cells in this element
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_ipVolume(i,e) = math_areaTriangle(mesh_cellnode(1:3,mesh_cell(1,i,e)), & ! here we assume a planar shape, so division in two triangles suffices
mesh_cellnode(1:3,mesh_cell(2,i,e)), &
mesh_cellnode(1:3,mesh_cell(3,i,e))) &
+ math_areaTriangle(mesh_cellnode(1:3,mesh_cell(3,i,e)), &
mesh_cellnode(1:3,mesh_cell(4,i,e)), &
mesh_cellnode(1:3,mesh_cell(1,i,e)))
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
case (3_pInt) ! 3D 4node
forall (i = 1_pInt:FE_Nips(g)) & ! loop over ips=cells in this element
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_ipVolume(i,e) = math_volTetrahedron(mesh_cellnode(1:3,mesh_cell(1,i,e)), &
mesh_cellnode(1:3,mesh_cell(2,i,e)), &
mesh_cellnode(1:3,mesh_cell(3,i,e)), &
mesh_cellnode(1:3,mesh_cell(4,i,e)))
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
case (4_pInt) ! 3D 8node
m = FE_NcellnodesPerCellface(c)
do i = 1_pInt,FE_Nips(g) ! loop over ips=cells in this element
subvolume = 0.0_pReal
forall(f = 1_pInt:FE_NipNeighbors(c), n = 1_pInt:FE_NcellnodesPerCellface(c)) &
subvolume(n,f) = math_volTetrahedron(&
calculation of cellnode positions now available in python via damask.core first call damask.core.mesh.mesh_init_postprocessing(meshfilename) to initialize all necessary mesh variables then damask.core.mesh.mesh_build_cellnodes(nodes) calculates the cellnode positions for a given list of node positions the meshfile that is needed for the init is created automatically by mesh_init in DAMASK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-22 00:18:59 +05:30
mesh_cellnode(1:3,mesh_cell(FE_cellface( n ,f,c),i,e)), &
mesh_cellnode(1:3,mesh_cell(FE_cellface(1+mod(n ,m),f,c),i,e)), &
mesh_cellnode(1:3,mesh_cell(FE_cellface(1+mod(n+1,m),f,c),i,e)), &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
mesh_ipCoordinates(1:3,i,e))
mesh_ipVolume(i,e) = 0.5_pReal * sum(subvolume) ! each subvolume is based on four tetrahedrons, altough the face consists of only two triangles -> averaging factor two
2d elements now have a volume corresponding to a thickness of 1; used to have zero volume now also 2d elements can be used with nonlocal model
2013-04-09 23:37:30 +05:30
enddo
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
end select
enddo
!$OMP END PARALLEL DO
Now able to have real periodicity for fluxes (fluxes leaving the model on one side enter on the other side). To enable this feature one has to add the following somewhere in the marc input file: $mpie periodic x y z for having periodicity in all directions $mpie periodic z x for having periodicity in x and z direction etc. Note that this only works for regular meshes!!!
2011-02-16 21:53:08 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end subroutine mesh_build_ipVolumes
Now able to have real periodicity for fluxes (fluxes leaving the model on one side enter on the other side). To enable this feature one has to add the following somewhere in the marc input file: $mpie periodic x y z for having periodicity in all directions $mpie periodic z x for having periodicity in x and z direction etc. Note that this only works for regular meshes!!!
2011-02-16 21:53:08 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
introduced distinction between ip coordinates and cell center coordinates; the former can be globally calculated by subroutine mesh_build_ipCoordinates, the latter locally by the function mesh_cellCenterCoordinates; renamed mesh_ipCenterOfGravity to mesh_ipCordinates
2012-11-06 20:07:13 +05:30
!> @brief Calculates IP Coordinates. Allocates global array 'mesh_ipCoordinates'
! Called by all solvers in mesh_init in order to initialize the ip coordinates.
! Later on the current ip coordinates are directly prvided by the spectral solver and by Abaqus,
! so no need to use this subroutine anymore; Marc however only provides nodal displacements,
! so in this case the ip coordinates are always calculated on the basis of this subroutine.
! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
! FOR THE MOMENT THIS SUBROUTINE ACTUALLY CALCULATES THE CELL CENTER AND NOT THE IP COORDINATES,
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
! AS THE IP IS NOT (ALWAYS) LOCATED IN THE CENTER OF THE IP VOLUME.
! HAS TO BE CHANGED IN A LATER VERSION.
introduced distinction between ip coordinates and cell center coordinates; the former can be globally calculated by subroutine mesh_build_ipCoordinates, the latter locally by the function mesh_cellCenterCoordinates; renamed mesh_ipCenterOfGravity to mesh_ipCordinates
2012-11-06 20:07:13 +05:30
! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
subroutine mesh_build_ipCoordinates
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
integer(pInt) :: e,t,g,c,i,n
real(pReal), dimension(3) :: myCoords
Now able to have real periodicity for fluxes (fluxes leaving the model on one side enter on the other side). To enable this feature one has to add the following somewhere in the marc input file: $mpie periodic x y z for having periodicity in all directions $mpie periodic z x for having periodicity in x and z direction etc. Note that this only works for regular meshes!!!
2011-02-16 21:53:08 +05:30
compiler complained
2017-04-25 16:04:14 +05:30
if (.not. allocated(mesh_ipCoordinates)) &
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
allocate(mesh_ipCoordinates(3,theMesh%elem%nIPs,mesh_NcpElems),source=0.0_pReal)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!$OMP PARALLEL DO PRIVATE(t,g,c,myCoords)
compiler complained
2017-04-25 16:04:14 +05:30
do e = 1_pInt,mesh_NcpElems ! loop over cpElems
t = mesh_element(2_pInt,e) ! get element type
g = FE_geomtype(t) ! get geometry type
c = FE_celltype(g) ! get cell type
do i = 1_pInt,FE_Nips(g) ! loop over ips=cells in this element
myCoords = 0.0_pReal
do n = 1_pInt,FE_NcellnodesPerCell(c) ! loop over cell nodes in this cell
myCoords = myCoords + mesh_cellnode(1:3,mesh_cell(n,i,e))
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
enddo
compiler complained
2017-04-25 16:04:14 +05:30
mesh_ipCoordinates(1:3,i,e) = myCoords / real(FE_NcellnodesPerCell(c),pReal)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
enddo
compiler complained
2017-04-25 16:04:14 +05:30
enddo
parallelized mesh_build_subNodeCoords and mesh_build_ipCoordinates
2012-11-09 13:17:14 +05:30
!$OMP END PARALLEL DO
Now able to have real periodicity for fluxes (fluxes leaving the model on one side enter on the other side). To enable this feature one has to add the following somewhere in the marc input file: $mpie periodic x y z for having periodicity in all directions $mpie periodic z x for having periodicity in x and z direction etc. Note that this only works for regular meshes!!!
2011-02-16 21:53:08 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end subroutine mesh_build_ipCoordinates
added _build_ipNeighborhood and _faceMatch
2007-03-26 14:20:57 +05:30
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
introduced distinction between ip coordinates and cell center coordinates; the former can be globally calculated by subroutine mesh_build_ipCoordinates, the latter locally by the function mesh_cellCenterCoordinates; renamed mesh_ipCenterOfGravity to mesh_ipCordinates
2012-11-06 20:07:13 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief Calculates cell center coordinates.
!--------------------------------------------------------------------------------------------------
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
pure function mesh_cellCenterCoordinates(ip,el)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
implicit none
integer(pInt), intent(in) :: el, & !< element number
ip !< integration point number
real(pReal), dimension(3) :: mesh_cellCenterCoordinates !< x,y,z coordinates of the cell center of the requested IP cell
integer(pInt) :: t,g,c,n
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
t = mesh_element(2_pInt,el) ! get element type
g = FE_geomtype(t) ! get geometry type
c = FE_celltype(g) ! get cell type
mesh_cellCenterCoordinates = 0.0_pReal
do n = 1_pInt,FE_NcellnodesPerCell(c) ! loop over cell nodes in this cell
mesh_cellCenterCoordinates = mesh_cellCenterCoordinates + mesh_cellnode(1:3,mesh_cell(n,ip,el))
enddo
compiler complained
2017-04-25 16:04:14 +05:30
mesh_cellCenterCoordinates = mesh_cellCenterCoordinates / real(FE_NcellnodesPerCell(c),pReal)
introduced distinction between ip coordinates and cell center coordinates; the former can be globally calculated by subroutine mesh_build_ipCoordinates, the latter locally by the function mesh_cellCenterCoordinates; renamed mesh_ipCenterOfGravity to mesh_ipCordinates
2012-11-06 20:07:13 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
end function mesh_cellCenterCoordinates
introduced distinction between ip coordinates and cell center coordinates; the former can be globally calculated by subroutine mesh_build_ipCoordinates, the latter locally by the function mesh_cellCenterCoordinates; renamed mesh_ipCenterOfGravity to mesh_ipCordinates
2012-11-06 20:07:13 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
!> @brief Parses geometry file
bugfixes: theMesh%Nelems need to be set (using an intermediate function until a routine does that) spectral.geom file can have "N+n to N" (backwards counting)
2019-02-02 16:16:38 +05:30
!> @details important variables have an implicit "save" attribute. Therefore, this function is
! supposed to be called only once!
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
subroutine mesh_spectral_read_grid()
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
use IO, only: &
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
IO_stringPos, &
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
IO_lc, &
IO_stringValue, &
IO_intValue, &
IO_floatValue, &
IO_error
use DAMASK_interface, only: &
geometryFile
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
implicit none
character(len=:), allocatable :: rawData
character(len=65536) :: line
integer(pInt), allocatable, dimension(:) :: chunkPos
integer(pInt), dimension(3) :: g = -1_pInt
real(pReal), dimension(3) :: s = -1_pInt
integer(pInt) :: h =- 1_pInt
integer(pInt) :: &
bugfixes: theMesh%Nelems need to be set (using an intermediate function until a routine does that) spectral.geom file can have "N+n to N" (backwards counting)
2019-02-02 16:16:38 +05:30
headerLength = -1_pInt, & !< length of header (in lines)
fileLength, & !< lenght of the geom file (in characters)
only parse geom file once
2019-02-02 13:48:01 +05:30
fileUnit, &
startPos, endPos, &
myStat, &
bugfixes: theMesh%Nelems need to be set (using an intermediate function until a routine does that) spectral.geom file can have "N+n to N" (backwards counting)
2019-02-02 16:16:38 +05:30
l, & !< line counter
c, & !< counter for # microstructures in line
o, & !< order of "to" packing
e, & !< "element", i.e. spectral collocation point
i, j
only parse geom file once
2019-02-02 13:48:01 +05:30
logical :: &
gotGrid = .false., &
gotSize = .false., &
gotHomogenization = .false.
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
! read data as stream
inquire(file = trim(geometryFile), size=fileLength)
open(newunit=fileUnit, file=trim(geometryFile), access='stream',&
status='old', position='rewind', action='read',iostat=myStat)
if(myStat /= 0_pInt) call IO_error(100_pInt,ext_msg=trim(geometryFile))
allocate(character(len=fileLength)::rawData)
read(fileUnit) rawData
close(fileUnit)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
! get header length
endPos = index(rawData,new_line(''))
if(endPos <= index(rawData,'head')) then
call IO_error(error_ID=841_pInt, ext_msg='mesh_spectral_read_grid')
else
chunkPos = IO_stringPos(rawData(1:endPos))
if (chunkPos(1) < 2_pInt) call IO_error(error_ID=841_pInt, ext_msg='mesh_spectral_read_grid')
headerLength = IO_intValue(rawData(1:endPos),chunkPos,1_pInt)
startPos = endPos + 1_pInt
endif
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
!--------------------------------------------------------------------------------------------------
! read and interprete header
l = 0
do while (l < headerLength .and. startPos < len(rawData))
endPos = startPos + index(rawData(startPos:),new_line('')) - 1_pInt
line = rawData(startPos:endPos)
startPos = endPos + 1_pInt
l = l + 1_pInt
! cycle empty lines
chunkPos = IO_stringPos(trim(line))
select case ( IO_lc(IO_StringValue(trim(line),chunkPos,1_pInt,.true.)) )
case ('grid')
if (chunkPos(1) > 6) gotGrid = .true.
do j = 2_pInt,6_pInt,2_pInt
select case (IO_lc(IO_stringValue(line,chunkPos,j)))
case('a')
g(1) = IO_intValue(line,chunkPos,j+1_pInt)
case('b')
g(2) = IO_intValue(line,chunkPos,j+1_pInt)
case('c')
g(3) = IO_intValue(line,chunkPos,j+1_pInt)
end select
enddo
case ('size')
if (chunkPos(1) > 6) gotSize = .true.
do j = 2_pInt,6_pInt,2_pInt
select case (IO_lc(IO_stringValue(line,chunkPos,j)))
case('x')
s(1) = IO_floatValue(line,chunkPos,j+1_pInt)
case('y')
s(2) = IO_floatValue(line,chunkPos,j+1_pInt)
case('z')
s(3) = IO_floatValue(line,chunkPos,j+1_pInt)
end select
enddo
case ('homogenization')
if (chunkPos(1) > 1) gotHomogenization = .true.
h = IO_intValue(line,chunkPos,2_pInt)
end select
enddo
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
!--------------------------------------------------------------------------------------------------
! global data
grid = g
geomSize = s
allocate(microGlobal(product(grid)), source = -1_pInt)
!--------------------------------------------------------------------------------------------------
! read and interprete content
e = 1_pInt
do while (startPos < len(rawData))
endPos = startPos + index(rawData(startPos:),new_line('')) - 1_pInt
line = rawData(startPos:endPos)
startPos = endPos + 1_pInt
l = l + 1_pInt
chunkPos = IO_stringPos(trim(line))
if (chunkPos(1) == 3) then
if (IO_lc(IO_stringValue(line,chunkPos,2)) == 'of') then
c = IO_intValue(line,chunkPos,1)
microGlobal(e:e+c-1_pInt) = [(IO_intValue(line,chunkPos,3),i = 1_pInt,IO_intValue(line,chunkPos,1))]
else if (IO_lc(IO_stringValue(line,chunkPos,2)) == 'to') then
bugfixes: theMesh%Nelems need to be set (using an intermediate function until a routine does that) spectral.geom file can have "N+n to N" (backwards counting)
2019-02-02 16:16:38 +05:30
c = abs(IO_intValue(line,chunkPos,3) - IO_intValue(line,chunkPos,1)) + 1_pInt
o = merge(+1_pInt, -1_pInt, IO_intValue(line,chunkPos,3) > IO_intValue(line,chunkPos,1))
microGlobal(e:e+c-1_pInt) = [(i, i = IO_intValue(line,chunkPos,1),IO_intValue(line,chunkPos,3),o)]
only parse geom file once
2019-02-02 13:48:01 +05:30
else
c = chunkPos(1)
do i = 0_pInt, c - 1_pInt
microGlobal(e+i) = IO_intValue(line,chunkPos,i+1_pInt)
enddo
endif
else
c = chunkPos(1)
do i = 0_pInt, c - 1_pInt
microGlobal(e+i) = IO_intValue(line,chunkPos,i+1_pInt)
enddo
endif
e = e+c
end do
if (e-1 /= product(grid)) print*, 'mist', e
! if (.not. gotGrid) &
! call IO_error(error_ID = 845_pInt, ext_msg='grid')
! if(any(mesh_spectral_getGrid < 1_pInt)) &
! call IO_error(error_ID = 843_pInt, ext_msg='mesh_spectral_getGrid')
! if (.not. gotSize) &
! call IO_error(error_ID = 845_pInt, ext_msg='size')
! if (any(mesh_spectral_getSize<=0.0_pReal)) &
! call IO_error(error_ID = 844_pInt, ext_msg='mesh_spectral_getSize')
! if (.not. gotHomogenization ) &
! call IO_error(error_ID = 845_pInt, ext_msg='homogenization')
! if (mesh_spectral_getHomogenization<1_pInt) &
! call IO_error(error_ID = 842_pInt, ext_msg='mesh_spectral_getHomogenization')
end subroutine mesh_spectral_read_grid
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
!> @brief Reads homogenization information from geometry file.
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
integer(pInt) function mesh_spectral_getHomogenization()
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
use IO, only: &
IO_checkAndRewind, &
IO_open_file, &
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
IO_stringPos, &
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
IO_lc, &
IO_stringValue, &
IO_intValue, &
IO_error
use DAMASK_interface, only: &
geometryFile
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
integer(pInt), allocatable, dimension(:) :: chunkPos
enabled alternative (and soon standard) keywords grid (resolution) and size (dimension). removed errors for odd resolution
2013-04-08 19:52:32 +05:30
integer(pInt) :: headerLength = 0_pInt
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
character(len=1024) :: line, &
keyword
added 'reset' flag for recursive function IO_read, need to reset stack when stopping exectution (will be used by constitutive and homogenization)
2013-12-11 22:19:20 +05:30
integer(pInt) :: i, myFileUnit
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
logical :: gotHomogenization = .false.
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
added 'reset' flag for recursive function IO_read, need to reset stack when stopping exectution (will be used by constitutive and homogenization)
2013-12-11 22:19:20 +05:30
myFileUnit = 289_pInt
call IO_open_file(myFileUnit,trim(geometryFile))
only parse geom file once
2019-02-02 13:48:01 +05:30
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
added 'reset' flag for recursive function IO_read, need to reset stack when stopping exectution (will be used by constitutive and homogenization)
2013-12-11 22:19:20 +05:30
call IO_checkAndRewind(myFileUnit)
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
added 'reset' flag for recursive function IO_read, need to reset stack when stopping exectution (will be used by constitutive and homogenization)
2013-12-11 22:19:20 +05:30
read(myFileUnit,'(a1024)') line
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
chunkPos = IO_stringPos(line)
keyword = IO_lc(IO_StringValue(line,chunkPos,2_pInt,.true.))
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
if (keyword(1:4) == 'head') then
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
headerLength = IO_intValue(line,chunkPos,1_pInt) + 1_pInt
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
else
call IO_error(error_ID=841_pInt, ext_msg='mesh_spectral_getHomogenization')
endif
added 'reset' flag for recursive function IO_read, need to reset stack when stopping exectution (will be used by constitutive and homogenization)
2013-12-11 22:19:20 +05:30
rewind(myFileUnit)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
do i = 1_pInt, headerLength
added 'reset' flag for recursive function IO_read, need to reset stack when stopping exectution (will be used by constitutive and homogenization)
2013-12-11 22:19:20 +05:30
read(myFileUnit,'(a1024)') line
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
chunkPos = IO_stringPos(line)
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
select case ( IO_lc(IO_StringValue(line,chunkPos,1,.true.)) )
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
case ('homogenization')
gotHomogenization = .true.
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name. functions in the core module are not used, including kdTree
2015-08-28 13:08:48 +05:30
mesh_spectral_getHomogenization = IO_intValue(line,chunkPos,2_pInt)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end select
enddo
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
close(myFileUnit)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
if (.not. gotHomogenization ) &
call IO_error(error_ID = 845_pInt, ext_msg='homogenization')
if (mesh_spectral_getHomogenization<1_pInt) &
call IO_error(error_ID = 842_pInt, ext_msg='mesh_spectral_getHomogenization')
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end function mesh_spectral_getHomogenization
improved reading in of values, now only warnings in case of problematic entries in material.config divergence calculation sqrt scaling optionally introduced for basic scheme spectral solver
2013-02-08 21:25:53 +05:30
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief Store x,y,z coordinates of all nodes in mesh.
!! Allocates global arrays 'mesh_node0' and 'mesh_node'
!--------------------------------------------------------------------------------------------------
took some gfortran complaints serious (unused imports, implicit castings)
2015-04-11 17:17:33 +05:30
subroutine mesh_spectral_build_nodes()
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
implicit none
took some gfortran complaints serious (unused imports, implicit castings)
2015-04-11 17:17:33 +05:30
integer(pInt) :: n
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
allocate (mesh_node0 (3,mesh_Nnodes), source = 0.0_pReal)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
more patches to get the tests running again: - fixed increment counting in postResults to current output scheme - corrected node coordinates calculation - some work on restart (restart test still not running)
2015-03-27 02:49:28 +05:30
forall (n = 0_pInt:mesh_Nnodes-1_pInt)
mesh_node0(1,n+1_pInt) = mesh_unitlength * &
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
geomSize(1)*real(mod(n,(grid(1)+1_pInt) ),pReal) &
/ real(grid(1),pReal)
more patches to get the tests running again: - fixed increment counting in postResults to current output scheme - corrected node coordinates calculation - some work on restart (restart test still not running)
2015-03-27 02:49:28 +05:30
mesh_node0(2,n+1_pInt) = mesh_unitlength * &
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
geomSize(2)*real(mod(n/(grid(1)+1_pInt),(grid(2)+1_pInt)),pReal) &
/ real(grid(2),pReal)
more patches to get the tests running again: - fixed increment counting in postResults to current output scheme - corrected node coordinates calculation - some work on restart (restart test still not running)
2015-03-27 02:49:28 +05:30
mesh_node0(3,n+1_pInt) = mesh_unitlength * &
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
size3*real(mod(n/(grid(1)+1_pInt)/(grid(2)+1_pInt),(grid3+1_pInt)),pReal) &
/ real(grid3,pReal) + &
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
size3offset
end forall
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
mesh_node = mesh_node0
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
end subroutine mesh_spectral_build_nodes
!--------------------------------------------------------------------------------------------------
!> @brief Store FEid, type, material, texture, and node list per element.
!! Allocates global array 'mesh_element'
made doxygen working for MSC.Marc again, small changes on the todo-statements for doxygen, they don't work on single lines of codes but only on module/variables
2013-06-11 22:05:04 +05:30
!> @todo does the IO_error makes sense?
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
!--------------------------------------------------------------------------------------------------
only parse geom file once
2019-02-02 13:48:01 +05:30
subroutine mesh_spectral_build_elements()
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
use IO, only: &
only parse geom file once
2019-02-02 13:48:01 +05:30
IO_error
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
implicit none
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
integer(pInt) :: &
e, i, &
only parse geom file once
2019-02-02 13:48:01 +05:30
some small iprovements: reading in geometry in mesh, readability of math, output formatting in fesolving and debug, hdf5 in constitutive
2014-05-15 14:22:16 +05:30
homog, &
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
elemOffset
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
only parse geom file once
2019-02-02 13:48:01 +05:30
homog = mesh_spectral_getHomogenization()
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
better name
2018-09-23 23:28:43 +05:30
allocate(mesh_element (4_pInt+8_pInt,mesh_NcpElems), source = 0_pInt)
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
initialize mesh and element
2019-02-01 16:54:23 +05:30
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
elemOffset = product(grid(1:2))*grid3Offset
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
e = 0_pInt
do while (e < mesh_NcpElems) ! fill expected number of elements, stop at end of data (or blank line!)
e = e+1_pInt ! valid element entry
mesh_element(1,:) only used for debug output set to -1 at the moment to indicate that it is not used. Re-implementation should be done for MSC.Marc and Abaqus only.
2018-09-23 22:10:14 +05:30
mesh_element( 1,e) = -1_pInt ! DEPRECATED
initialize mesh and element
2019-02-01 16:54:23 +05:30
mesh_element( 2,e) = 10_pInt
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
mesh_element( 3,e) = homog ! homogenization
more descriptive names
2018-09-23 22:20:54 +05:30
mesh_element( 4,e) = microGlobal(e+elemOffset) ! microstructure
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_element( 5,e) = e + (e-1_pInt)/grid(1) + &
((e-1_pInt)/(grid(1)*grid(2)))*(grid(1)+1_pInt) ! base node
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
mesh_element( 6,e) = mesh_element(5,e) + 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_element( 7,e) = mesh_element(5,e) + grid(1) + 2_pInt
mesh_element( 8,e) = mesh_element(5,e) + grid(1) + 1_pInt
mesh_element( 9,e) = mesh_element(5,e) +(grid(1) + 1_pInt) * (grid(2) + 1_pInt) ! second floor base node
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
mesh_element(10,e) = mesh_element(9,e) + 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_element(11,e) = mesh_element(9,e) + grid(1) + 2_pInt
mesh_element(12,e) = mesh_element(9,e) + grid(1) + 1_pInt
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 21:36:19 +05:30
enddo
made doxygen working for MSC.Marc again, small changes on the todo-statements for doxygen, they don't work on single lines of codes but only on module/variables
2013-06-11 22:05:04 +05:30
if (e /= mesh_NcpElems) call IO_error(880_pInt,e)
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
end subroutine mesh_spectral_build_elements
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
spectral_build_neighborhood now "hidden" in spectral compile part. (should not interfere with Marc/Abaqus compile anymore)
2013-04-22 14:31:58 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief build neighborhood relations for spectral
!> @details assign globals: mesh_ipNeighborhood
!--------------------------------------------------------------------------------------------------
removed a bunch of unneded functions
2016-06-27 21:45:56 +05:30
subroutine mesh_spectral_build_ipNeighborhood
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
implicit none
integer(pInt) :: &
x,y,z, &
e
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
allocate(mesh_ipNeighborhood(3,mesh_maxNipNeighbors,theMesh%elem%nIPs,mesh_NcpElems),source=0_pInt)
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
e = 0_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
do z = 0_pInt,grid3-1_pInt
do y = 0_pInt,grid(2)-1_pInt
do x = 0_pInt,grid(1)-1_pInt
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
e = e + 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_ipNeighborhood(1,1,1,e) = z * grid(1) * grid(2) &
+ y * grid(1) &
+ modulo(x+1_pInt,grid(1)) &
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
+ 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_ipNeighborhood(1,2,1,e) = z * grid(1) * grid(2) &
+ y * grid(1) &
+ modulo(x-1_pInt,grid(1)) &
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
+ 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_ipNeighborhood(1,3,1,e) = z * grid(1) * grid(2) &
+ modulo(y+1_pInt,grid(2)) * grid(1) &
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
+ x &
+ 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_ipNeighborhood(1,4,1,e) = z * grid(1) * grid(2) &
+ modulo(y-1_pInt,grid(2)) * grid(1) &
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
+ x &
+ 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_ipNeighborhood(1,5,1,e) = modulo(z+1_pInt,grid3) * grid(1) * grid(2) &
+ y * grid(1) &
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
+ x &
+ 1_pInt
clearer naming, debug options for spectral do not work for MPI
2015-09-11 14:22:03 +05:30
mesh_ipNeighborhood(1,6,1,e) = modulo(z-1_pInt,grid3) * grid(1) * grid(2) &
+ y * grid(1) &
moved public data res,size and homog from mesh to DAMASK_spectral_utilities (as grid and geomSize)
2013-05-08 21:22:29 +05:30
+ x &
+ 1_pInt
mesh_ipNeighborhood(2,1:6,1,e) = 1_pInt
mesh_ipNeighborhood(3,1,1,e) = 2_pInt
mesh_ipNeighborhood(3,2,1,e) = 1_pInt
mesh_ipNeighborhood(3,3,1,e) = 4_pInt
mesh_ipNeighborhood(3,4,1,e) = 3_pInt
mesh_ipNeighborhood(3,5,1,e) = 6_pInt
mesh_ipNeighborhood(3,6,1,e) = 5_pInt
enddo
enddo
enddo
spectral_build_neighborhood now "hidden" in spectral compile part. (should not interfere with Marc/Abaqus compile anymore)
2013-04-22 14:31:58 +05:30
end subroutine mesh_spectral_build_ipNeighborhood
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief builds mesh of (distorted) cubes for given coordinates (= center of the cubes)
!--------------------------------------------------------------------------------------------------
function mesh_nodesAroundCentres(gDim,Favg,centres) result(nodes)
use debug, only: &
debug_mesh, &
debug_level, &
debug_levelBasic
use math, only: &
math_mul33x3
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
real(pReal), intent(in), dimension(:,:,:,:) :: &
centres
real(pReal), dimension(3,size(centres,2)+1,size(centres,3)+1,size(centres,4)+1) :: &
nodes
real(pReal), intent(in), dimension(3) :: &
gDim
real(pReal), intent(in), dimension(3,3) :: &
Favg
real(pReal), dimension(3,size(centres,2)+2,size(centres,3)+2,size(centres,4)+2) :: &
wrappedCentres
integer(pInt) :: &
i,j,k,n
integer(pInt), dimension(3), parameter :: &
diag = 1_pInt
integer(pInt), dimension(3) :: &
shift = 0_pInt, &
lookup = 0_pInt, &
me = 0_pInt, &
iRes = 0_pInt
integer(pInt), dimension(3,8) :: &
neighbor = reshape([ &
0_pInt, 0_pInt, 0_pInt, &
1_pInt, 0_pInt, 0_pInt, &
1_pInt, 1_pInt, 0_pInt, &
0_pInt, 1_pInt, 0_pInt, &
0_pInt, 0_pInt, 1_pInt, &
1_pInt, 0_pInt, 1_pInt, &
1_pInt, 1_pInt, 1_pInt, &
0_pInt, 1_pInt, 1_pInt ], [3,8])
!--------------------------------------------------------------------------------------------------
! initializing variables
iRes = [size(centres,2),size(centres,3),size(centres,4)]
nodes = 0.0_pReal
wrappedCentres = 0.0_pReal
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
!--------------------------------------------------------------------------------------------------
! report
if (iand(debug_level(debug_mesh),debug_levelBasic) /= 0_pInt) then
write(6,'(a)') ' Meshing cubes around centroids'
write(6,'(a,3(e12.5))') ' Dimension: ', gDim
write(6,'(a,3(i5))') ' Resolution:', iRes
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
endif
changed assumed arrays (:) to properly defined ones (gfortran 4.7 was complaining) and remove use statements that are not needed
2012-02-21 21:09:36 +05:30
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
!--------------------------------------------------------------------------------------------------
! building wrappedCentres = centroids + ghosts
wrappedCentres(1:3,2_pInt:iRes(1)+1_pInt,2_pInt:iRes(2)+1_pInt,2_pInt:iRes(3)+1_pInt) = centres
do k = 0_pInt,iRes(3)+1_pInt
do j = 0_pInt,iRes(2)+1_pInt
do i = 0_pInt,iRes(1)+1_pInt
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
if (k==0_pInt .or. k==iRes(3)+1_pInt .or. & ! z skin
j==0_pInt .or. j==iRes(2)+1_pInt .or. & ! y skin
i==0_pInt .or. i==iRes(1)+1_pInt ) then ! x skin
me = [i,j,k] ! me on skin
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
shift = sign(abs(iRes+diag-2_pInt*me)/(iRes+diag),iRes+diag-2_pInt*me)
lookup = me-diag+shift*iRes
wrappedCentres(1:3,i+1_pInt, j+1_pInt, k+1_pInt) = &
compiler complained
2017-04-25 16:04:14 +05:30
centres(1:3,lookup(1)+1_pInt,lookup(2)+1_pInt,lookup(3)+1_pInt) &
- math_mul33x3(Favg, real(shift,pReal)*gDim)
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
endif
enddo; enddo; enddo
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
!--------------------------------------------------------------------------------------------------
! averaging
do k = 0_pInt,iRes(3); do j = 0_pInt,iRes(2); do i = 0_pInt,iRes(1)
do n = 1_pInt,8_pInt
nodes(1:3,i+1_pInt,j+1_pInt,k+1_pInt) = &
nodes(1:3,i+1_pInt,j+1_pInt,k+1_pInt) + wrappedCentres(1:3,i+1_pInt+neighbor(1,n), &
j+1_pInt+neighbor(2,n), &
k+1_pInt+neighbor(3,n) )
enddo
enddo; enddo; enddo
moved some more 'mesh related' functions for post processing from math.f90 to mesh.f90 f2py functions remaining in math.f90 now uses assumed size arrays in order to have simpler interfaces. This is only working with python 2.7! changed python pre- and postprocessing scripts. If you encounter any problems whith core modules, try to remove the old core.so in the lib/damask
2012-08-27 13:34:47 +05:30
nodes = nodes/8.0_pReal
changed all remaining routines to fortran-fast arrays (geometry reconstruction etc.) changed all remaining routines in f2py to more clever determination of array size (requires f2py >= 2.0) enabled 3D visualize to work with odd resolution by switching to linear reconstruction PLEASE NOTE: Redefinition of routines for f2py might cause trouble -> DELETE DAMASK_ROOT/lib/damask/core.so in this case further changes: added pure statement where possible, polished, unified use of "Q" for "Quaternion" and reordered math to have similar routines together
2013-01-31 21:58:08 +05:30
end function mesh_nodesAroundCentres
new errors regarding Spectral method input file parsing mesh setup from Spectral input file $Id$ in FEsolving activated
2010-05-06 22:10:47 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
!> @brief calculation of IP interface areas, allocate globals '_ipArea', and '_ipAreaNormal'
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!--------------------------------------------------------------------------------------------------
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
subroutine mesh_build_ipAreas
use math, only: &
math_crossproduct
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
added function calls in init()
2007-04-10 16:52:53 +05:30
implicit none
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
integer(pInt) :: e,t,g,c,i,f,n,m
real(pReal), dimension (3,FE_maxNcellnodesPerCellface) :: nodePos, normals
real(pReal), dimension(3) :: normal
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
avoiding duplicated variables
2019-02-02 17:01:07 +05:30
allocate(mesh_ipArea(mesh_maxNipNeighbors,theMesh%elem%nIPs,mesh_NcpElems), source=0.0_pReal)
allocate(mesh_ipAreaNormal(3_pInt,mesh_maxNipNeighbors,theMesh%elem%nIPs,mesh_NcpElems), source=0.0_pReal)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
!$OMP PARALLEL DO PRIVATE(t,g,c,nodePos,normal,normals)
do e = 1_pInt,mesh_NcpElems ! loop over cpElems
t = mesh_element(2_pInt,e) ! get element type
g = FE_geomtype(t) ! get geometry type
c = FE_celltype(g) ! get cell type
select case (c)
major debugging and streamlining of the calculation of ipVolumes and ipAreas
2009-01-20 00:12:31 +05:30
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
case (1_pInt,2_pInt) ! 2D 3 or 4 node
do i = 1_pInt,FE_Nips(g) ! loop over ips=cells in this element
do f = 1_pInt,FE_NipNeighbors(c) ! loop over cell faces
forall(n = 1_pInt:FE_NcellnodesPerCellface(c)) &
nodePos(1:3,n) = mesh_cellnode(1:3,mesh_cell(FE_cellface(n,f,c),i,e))
normal(1) = nodePos(2,2) - nodePos(2,1) ! x_normal = y_connectingVector
normal(2) = -(nodePos(1,2) - nodePos(1,1)) ! y_normal = -x_connectingVector
normal(3) = 0.0_pReal
mesh_ipArea(f,i,e) = norm2(normal)
mesh_ipAreaNormal(1:3,f,i,e) = normal / norm2(normal) ! ensure unit length of area normal
enddo
new Marc2017 file format finally working!
2018-02-02 19:36:13 +05:30
enddo
# new interface for Abaqus # IO has some additional functionality for Abaqus parsing # ping pong scheme in FE interface now similar (and more human understandable) in both versions # mesh has better splitting of different tasks, plus operation on database whenever possible # FEsolver as new global var to indicate FEM solver type # computation mode reshuffling: 6 is now Marc special case of recycling...
2009-10-12 21:31:49 +05:30
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
case (3_pInt) ! 3D 4node
do i = 1_pInt,FE_Nips(g) ! loop over ips=cells in this element
do f = 1_pInt,FE_NipNeighbors(c) ! loop over cell faces
forall(n = 1_pInt:FE_NcellnodesPerCellface(c)) &
nodePos(1:3,n) = mesh_cellnode(1:3,mesh_cell(FE_cellface(n,f,c),i,e))
normal = math_crossproduct(nodePos(1:3,2) - nodePos(1:3,1), &
nodePos(1:3,3) - nodePos(1:3,1))
mesh_ipArea(f,i,e) = norm2(normal)
mesh_ipAreaNormal(1:3,f,i,e) = normal / norm2(normal) ! ensure unit length of area normal
enddo
enddo
# new interface for Abaqus # IO has some additional functionality for Abaqus parsing # ping pong scheme in FE interface now similar (and more human understandable) in both versions # mesh has better splitting of different tasks, plus operation on database whenever possible # FEsolver as new global var to indicate FEM solver type # computation mode reshuffling: 6 is now Marc special case of recycling...
2009-10-12 21:31:49 +05:30
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
case (4_pInt) ! 3D 8node
! for this cell type we get the normal of the quadrilateral face as an average of
! four normals of triangular subfaces; since the face consists only of two triangles,
! the sum has to be divided by two; this whole prcedure tries to compensate for
! probable non-planar cell surfaces
m = FE_NcellnodesPerCellface(c)
do i = 1_pInt,FE_Nips(g) ! loop over ips=cells in this element
do f = 1_pInt,FE_NipNeighbors(c) ! loop over cell faces
forall(n = 1_pInt:FE_NcellnodesPerCellface(c)) &
nodePos(1:3,n) = mesh_cellnode(1:3,mesh_cell(FE_cellface(n,f,c),i,e))
forall(n = 1_pInt:FE_NcellnodesPerCellface(c)) &
normals(1:3,n) = 0.5_pReal &
* math_crossproduct(nodePos(1:3,1+mod(n ,m)) - nodePos(1:3,n), &
nodePos(1:3,1+mod(n+1,m)) - nodePos(1:3,n))
normal = 0.5_pReal * sum(normals,2)
mesh_ipArea(f,i,e) = norm2(normal)
mesh_ipAreaNormal(1:3,f,i,e) = normal / norm2(normal)
enddo
enddo
# new interface for Abaqus # IO has some additional functionality for Abaqus parsing # ping pong scheme in FE interface now similar (and more human understandable) in both versions # mesh has better splitting of different tasks, plus operation on database whenever possible # FEsolver as new global var to indicate FEM solver type # computation mode reshuffling: 6 is now Marc special case of recycling...
2009-10-12 21:31:49 +05:30
removing non-grid(spectral) related functionality
2019-01-24 14:42:27 +05:30
end select
enddo
!$OMP END PARALLEL DO
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
took some gfortran complaints serious (unused imports, implicit castings)
2015-04-11 17:17:33 +05:30
end subroutine mesh_build_ipAreas
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
improved regridding, especially interfacing to python (also for nearest neighbor search)
2012-04-24 22:32:27 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief get properties of different types of finite elements
!> @details assign globals: FE_nodesAtIP, FE_ipNeighbor, FE_cellnodeParentnodeWeights, FE_subNodeOnIPFace
!--------------------------------------------------------------------------------------------------
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
subroutine mesh_build_FEdata
merged code for python and spectral solver interfacing (shared most of it anyway). put functionality for getting header information (dimension, homogenization, resolution) in functions in mesh.f90
2012-04-11 22:58:08 +05:30
implicit none
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
integer(pInt) :: me
leaner syntax with sourced allocation
2018-09-23 20:56:13 +05:30
allocate(FE_nodesAtIP(FE_maxmaxNnodesAtIP,FE_maxNips,FE_Ngeomtypes), source=0_pInt)
allocate(FE_ipNeighbor(FE_maxNipNeighbors,FE_maxNips,FE_Ngeomtypes), source=0_pInt)
allocate(FE_cell(FE_maxNcellnodesPerCell,FE_maxNips,FE_Ngeomtypes), source=0_pInt)
allocate(FE_cellnodeParentnodeWeights(FE_maxNnodes,FE_maxNcellnodes,FE_Nelemtypes), source=0.0_pReal)
allocate(FE_cellface(FE_maxNcellnodesPerCellface,FE_maxNcellfaces,FE_Ncelltypes), source=0_pInt)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
!*** fill FE_nodesAtIP with data ***
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = 0_pInt
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 6 (2D 3node 1ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
1,2,3 &
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
several small corrections/polishing
2012-04-20 17:28:41 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 125 (2D 6node 3ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
1, &
2, &
3 &
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
fixed bug concerning file handling.
2012-04-12 00:16:36 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 11 (2D 4node 4ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
1, &
2, &
4, &
3 &
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
merged code for python and spectral solver interfacing (shared most of it anyway). put functionality for getting header information (dimension, homogenization, resolution) in functions in mesh.f90
2012-04-11 22:58:08 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 27 (2D 8node 9ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
1,0, &
1,2, &
2,0, &
1,4, &
0,0, &
2,3, &
4,0, &
3,4, &
3,0 &
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
merged code for python and spectral solver interfacing (shared most of it anyway). put functionality for getting header information (dimension, homogenization, resolution) in functions in mesh.f90
2012-04-11 22:58:08 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 134 (3D 4node 1ip)
reshape(int([&
1,2,3,4 &
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 127 (3D 10node 4ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
1, &
2, &
3, &
4 &
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
merged code for python and spectral solver interfacing (shared most of it anyway). put functionality for getting header information (dimension, homogenization, resolution) in functions in mesh.f90
2012-04-11 22:58:08 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 136 (3D 6node 6ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
1, &
2, &
3, &
4, &
5, &
6 &
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
several small corrections/polishing
2012-04-20 17:28:41 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 117 (3D 8node 1ip)
reshape(int([&
1,2,3,4,5,6,7,8 &
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 7 (3D 8node 8ip)
reshape(int([&
1, &
2, &
4, &
3, &
5, &
6, &
8, &
7 &
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
me = me + 1_pInt
FE_nodesAtIP(1:FE_maxNnodesAtIP(me),1:FE_Nips(me),me) = & ! element 21 (3D 20node 27ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
1,0, 0,0, &
1,2, 0,0, &
2,0, 0,0, &
1,4, 0,0, &
1,3, 2,4, &
2,3, 0,0, &
4,0, 0,0, &
3,4, 0,0, &
3,0, 0,0, &
1,5, 0,0, &
1,6, 2,5, &
2,6, 0,0, &
1,8, 4,5, &
0,0, 0,0, &
2,7, 3,6, &
4,8, 0,0, &
3,8, 4,7, &
3,7, 0,0, &
5,0, 0,0, &
5,6, 0,0, &
6,0, 0,0, &
5,8, 0,0, &
5,7, 6,8, &
6,7, 0,0, &
8,0, 0,0, &
7,8, 0,0, &
7,0, 0,0 &
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
],pInt),[FE_maxNnodesAtIP(me),FE_Nips(me)])
merged code for python and spectral solver interfacing (shared most of it anyway). put functionality for getting header information (dimension, homogenization, resolution) in functions in mesh.f90
2012-04-11 22:58:08 +05:30
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
! *** FE_ipNeighbor ***
! is a list of the neighborhood of each IP.
! It is sorted in (local) +x,-x, +y,-y, +z,-z direction.
! Positive integers denote an intra-FE IP identifier.
! Negative integers denote the interface behind which the neighboring (extra-FE) IP will be located.
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = 0_pInt
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 6 (2D 3node 1ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
-2,-3,-1 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 125 (2D 6node 3ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
2,-3, 3,-1, &
-2, 1, 3,-1, &
2,-3,-2, 1 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 11 (2D 4node 4ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
2,-4, 3,-1, &
-2, 1, 4,-1, &
4,-4,-3, 1, &
-2, 3,-3, 2 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 27 (2D 8node 9ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
2,-4, 4,-1, &
3, 1, 5,-1, &
-2, 2, 6,-1, &
5,-4, 7, 1, &
6, 4, 8, 2, &
-2, 5, 9, 3, &
8,-4,-3, 4, &
9, 7,-3, 5, &
-2, 8,-3, 6 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 134 (3D 4node 1ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
-1,-2,-3,-4 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 127 (3D 10node 4ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
2,-4, 3,-2, 4,-1, &
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
-2, 1, 3,-2, 4,-1, &
2,-4,-3, 1, 4,-1, &
2,-4, 3,-2,-3, 1 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 136 (3D 6node 6ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
2,-4, 3,-2, 4,-1, &
-3, 1, 3,-2, 5,-1, &
2,-4,-3, 1, 6,-1, &
5,-4, 6,-2,-5, 1, &
-3, 4, 6,-2,-5, 2, &
5,-4,-3, 4,-5, 3 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 117 (3D 8node 1ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
-3,-5,-4,-2,-6,-1 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 7 (3D 8node 8ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
2,-5, 3,-2, 5,-1, &
-3, 1, 4,-2, 6,-1, &
4,-5,-4, 1, 7,-1, &
-3, 3,-4, 2, 8,-1, &
6,-5, 7,-2,-6, 1, &
-3, 5, 8,-2,-6, 2, &
8,-5,-4, 5,-6, 3, &
-3, 7,-4, 6,-6, 4 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_ipNeighbor(1:FE_NipNeighbors(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 21 (3D 20node 27ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
2,-5, 4,-2,10,-1, &
3, 1, 5,-2,11,-1, &
-3, 2, 6,-2,12,-1, &
5,-5, 7, 1,13,-1, &
6, 4, 8, 2,14,-1, &
-3, 5, 9, 3,15,-1, &
8,-5,-4, 4,16,-1, &
9, 7,-4, 5,17,-1, &
-3, 8,-4, 6,18,-1, &
11,-5,13,-2,19, 1, &
12,10,14,-2,20, 2, &
-3,11,15,-2,21, 3, &
14,-5,16,10,22, 4, &
15,13,17,11,23, 5, &
-3,14,18,12,24, 6, &
17,-5,-4,13,25, 7, &
18,16,-4,14,26, 8, &
-3,17,-4,15,27, 9, &
20,-5,22,-2,-6,10, &
21,19,23,-2,-6,11, &
-3,20,24,-2,-6,12, &
23,-5,25,19,-6,13, &
24,22,26,20,-6,14, &
-3,23,27,21,-6,15, &
26,-5,-4,22,-6,16, &
27,25,-4,23,-6,17, &
-3,26,-4,24,-6,18 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pInt),[FE_NipNeighbors(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
! *** FE_cell ***
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = 0_pInt
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 6 (2D 3node 1ip)
reshape(int([&
1,2,3 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 125 (2D 6node 3ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 4, 7, 6, &
2, 5, 7, 4, &
3, 6, 7, 5 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 11 (2D 4node 4ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 5, 9, 8, &
5, 2, 6, 9, &
8, 9, 7, 4, &
9, 6, 3, 7 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 27 (2D 8node 9ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 5,13,12, &
5, 6,14,13, &
6, 2, 7,14, &
12,13,16,11, &
13,14,15,16, &
14, 7, 8,15, &
11,16,10, 4, &
16,15, 9,10, &
15, 8, 3, 9 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 134 (3D 4node 1ip)
reshape(int([&
1, 2, 3, 4 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 127 (3D 10node 4ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 5,11, 7, 8,12,15,14, &
5, 2, 6,11,12, 9,13,15, &
7,11, 6, 3,14,15,13,10, &
8,12,15, 4, 4, 9,13,10 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 136 (3D 6node 6ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 7,16, 9,10,17,21,19, &
7, 2, 8,16,17,11,18,21, &
9,16, 8, 3,19,21,18,12, &
10,17,21,19, 4,13,20,15, &
17,11,18,21,13, 5,14,20, &
19,21,18,12,15,20,14, 6 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 117 (3D 8node 1ip)
reshape(int([&
1, 2, 3, 4, 5, 6, 7, 8 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 7 (3D 8node 8ip)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 9,21,12,13,22,27,25, &
9, 2,10,21,22,14,23,27, &
12,21,11, 4,25,27,24,16, &
21,10, 3,11,27,23,15,24, &
13,22,27,25, 5,17,26,20, &
22,14,23,27,17, 6,18,26, &
25,27,24,16,20,26,19, 8, &
27,23,15,24,26,18, 7,19 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cell(1:FE_NcellnodesPerCell(FE_celltype(me)),1:FE_Nips(me),me) = & ! element 21 (3D 20node 27ip)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 9,33,16,17,37,57,44, &
9,10,34,33,37,38,58,57, &
10, 2,11,34,38,18,39,58, &
16,33,36,15,44,57,60,43, &
33,34,35,36,57,58,59,60, &
34,11,12,35,58,39,40,59, &
15,36,14, 4,43,60,42,20, &
36,35,13,14,60,59,41,42, &
35,12, 3,13,59,40,19,41, &
17,37,57,44,21,45,61,52, &
37,38,58,57,45,46,62,61, &
38,18,39,58,46,22,47,62, &
44,57,60,43,52,61,64,51, &
57,58,59,60,61,62,63,64, &
58,39,40,59,62,47,48,63, &
43,60,42,20,51,64,50,24, &
60,59,41,42,64,63,49,50, &
59,40,19,41,63,48,23,49, &
21,45,61,52, 5,25,53,32, &
45,46,62,61,25,26,54,53, &
46,22,47,62,26, 6,27,54, &
52,61,64,51,32,53,56,31, &
61,62,63,64,53,54,55,56, &
62,47,48,63,54,27,28,55, &
51,64,50,24,31,56,30, 8, &
64,63,49,50,56,55,29,30, &
63,48,23,49,55,28, 7,29 &
],pInt),[FE_NcellnodesPerCell(FE_celltype(me)),FE_Nips(me)])
! *** FE_cellnodeParentnodeWeights ***
! center of gravity of the weighted nodes gives the position of the cell node.
! fill with 0.
! example: face-centered cell node with face nodes 1,2,5,6 to be used in,
! e.g., an 8 node element, would be encoded:
! 1, 1, 0, 0, 1, 1, 0, 0
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = 0_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
me = me + 1_pInt
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 6 (2D 3node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, &
0, 1, 0, &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 1 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
me = me + 1_pInt
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 125 (2D 6node 3ip)
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, 0, 0, 0, &
0, 1, 0, 0, 0, 0, &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 1, 0, 0, 0, &
0, 0, 0, 1, 0, 0, &
0, 0, 0, 0, 1, 0, &
0, 0, 0, 0, 0, 1, &
1, 1, 1, 2, 2, 2 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 11 (2D 4node 4ip)
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, 0, &
0, 1, 0, 0, &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 1, 0, &
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
0, 0, 0, 1, &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 1, 0, 0, &
0, 1, 1, 0, &
0, 0, 1, 1, &
1, 0, 0, 1, &
1, 1, 1, 1 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 27 (2D 8node 9ip)
reshape(real([&
1, 0, 0, 0, 0, 0, 0, 0, &
0, 1, 0, 0, 0, 0, 0, 0, &
0, 0, 1, 0, 0, 0, 0, 0, &
0, 0, 0, 1, 0, 0, 0, 0, &
1, 0, 0, 0, 2, 0, 0, 0, &
0, 1, 0, 0, 2, 0, 0, 0, &
0, 1, 0, 0, 0, 2, 0, 0, &
0, 0, 1, 0, 0, 2, 0, 0, &
0, 0, 1, 0, 0, 0, 2, 0, &
0, 0, 0, 1, 0, 0, 2, 0, &
0, 0, 0, 1, 0, 0, 0, 2, &
1, 0, 0, 0, 0, 0, 0, 2, &
4, 1, 1, 1, 8, 2, 2, 8, &
1, 4, 1, 1, 8, 8, 2, 2, &
1, 1, 4, 1, 2, 8, 8, 2, &
1, 1, 1, 4, 2, 2, 8, 8 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 54 (2D 8node 4ip)
reshape(real([&
1, 0, 0, 0, 0, 0, 0, 0, &
0, 1, 0, 0, 0, 0, 0, 0, &
0, 0, 1, 0, 0, 0, 0, 0, &
0, 0, 0, 1, 0, 0, 0, 0, &
0, 0, 0, 0, 1, 0, 0, 0, &
0, 0, 0, 0, 0, 1, 0, 0, &
0, 0, 0, 0, 0, 0, 1, 0, &
0, 0, 0, 0, 0, 0, 0, 1, &
1, 1, 1, 1, 2, 2, 2, 2 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 134 (3D 4node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, 0, &
0, 1, 0, 0, &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 1, 0, &
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
0, 0, 0, 1 &
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 157 (3D 5node 4ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
reshape(real([&
1, 0, 0, 0, 0, &
0, 1, 0, 0, 0, &
0, 0, 1, 0, 0, &
0, 0, 0, 1, 0, &
1, 1, 0, 0, 0, &
0, 1, 1, 0, 0, &
1, 0, 1, 0, 0, &
1, 0, 0, 1, 0, &
0, 1, 0, 1, 0, &
0, 0, 1, 1, 0, &
1, 1, 1, 0, 0, &
1, 1, 0, 1, 0, &
0, 1, 1, 1, 0, &
1, 0, 1, 1, 0, &
0, 0, 0, 0, 1 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 127 (3D 10node 4ip)
reshape(real([&
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, &
0, 1, 0, 0, 0, 0, 0, 0, 0, 0, &
0, 0, 1, 0, 0, 0, 0, 0, 0, 0, &
0, 0, 0, 1, 0, 0, 0, 0, 0, 0, &
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, &
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, &
0, 0, 0, 0, 0, 0, 1, 0, 0, 0, &
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, &
0, 0, 0, 0, 0, 0, 0, 0, 1, 0, &
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, &
1, 1, 1, 0, 2, 2, 2, 0, 0, 0, &
1, 1, 0, 1, 2, 0, 0, 2, 2, 0, &
0, 1, 1, 1, 0, 2, 0, 0, 2, 2, &
1, 0, 1, 1, 0, 0, 2, 2, 0, 2, &
3, 3, 3, 3, 4, 4, 4, 4, 4, 4 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
me = me + 1_pInt
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 136 (3D 6node 6ip)
reshape(real([&
1, 0, 0, 0, 0, 0, &
0, 1, 0, 0, 0, 0, &
0, 0, 1, 0, 0, 0, &
0, 0, 0, 1, 0, 0, &
0, 0, 0, 0, 1, 0, &
0, 0, 0, 0, 0, 1, &
1, 1, 0, 0, 0, 0, &
0, 1, 1, 0, 0, 0, &
1, 0, 1, 0, 0, 0, &
1, 0, 0, 1, 0, 0, &
0, 1, 0, 0, 1, 0, &
0, 0, 1, 0, 0, 1, &
0, 0, 0, 1, 1, 0, &
0, 0, 0, 0, 1, 1, &
0, 0, 0, 1, 0, 1, &
1, 1, 1, 0, 0, 0, &
1, 1, 0, 1, 1, 0, &
0, 1, 1, 0, 1, 1, &
1, 0, 1, 1, 0, 1, &
0, 0, 0, 1, 1, 1, &
1, 1, 1, 1, 1, 1 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
me = me + 1_pInt
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 117 (3D 8node 1ip)
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
reshape(real([&
1, 0, 0, 0, 0, 0, 0, 0, &
0, 1, 0, 0, 0, 0, 0, 0, &
0, 0, 1, 0, 0, 0, 0, 0, &
0, 0, 0, 1, 0, 0, 0, 0, &
0, 0, 0, 0, 1, 0, 0, 0, &
0, 0, 0, 0, 0, 1, 0, 0, &
0, 0, 0, 0, 0, 0, 1, 0, &
0, 0, 0, 0, 0, 0, 0, 1 &
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
me = me + 1_pInt
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 7 (3D 8node 8ip)
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, 0, 0, 0, 0, 0, & !
0, 1, 0, 0, 0, 0, 0, 0, & !
0, 0, 1, 0, 0, 0, 0, 0, & !
0, 0, 0, 1, 0, 0, 0, 0, & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 0, 0, 1, 0, 0, 0, & ! 5
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
0, 0, 0, 0, 0, 1, 0, 0, & !
0, 0, 0, 0, 0, 0, 1, 0, & !
0, 0, 0, 0, 0, 0, 0, 1, & !
1, 1, 0, 0, 0, 0, 0, 0, & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 1, 1, 0, 0, 0, 0, 0, & ! 10
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
0, 0, 1, 1, 0, 0, 0, 0, & !
1, 0, 0, 1, 0, 0, 0, 0, & !
1, 0, 0, 0, 1, 0, 0, 0, & !
0, 1, 0, 0, 0, 1, 0, 0, & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 1, 0, 0, 0, 1, 0, & ! 15
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
0, 0, 0, 1, 0, 0, 0, 1, & !
0, 0, 0, 0, 1, 1, 0, 0, & !
0, 0, 0, 0, 0, 1, 1, 0, & !
0, 0, 0, 0, 0, 0, 1, 1, & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
0, 0, 0, 0, 1, 0, 0, 1, & ! 20
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 1, 1, 1, 0, 0, 0, 0, & !
1, 1, 0, 0, 1, 1, 0, 0, & !
0, 1, 1, 0, 0, 1, 1, 0, & !
0, 0, 1, 1, 0, 0, 1, 1, & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1, 0, 0, 1, 1, 0, 0, 1, & ! 25
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
0, 0, 0, 0, 1, 1, 1, 1, & !
1, 1, 1, 1, 1, 1, 1, 1 & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
me = me + 1_pInt
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 57 (3D 20node 8ip)
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & ! 5
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & ! 10
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, & ! 15
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, & ! 20
1, 1, 1, 1, 0, 0, 0, 0, 2, 2, 2, 2, 0, 0, 0, 0, 0, 0, 0, 0, & !
1, 1, 0, 0, 1, 1, 0, 0, 2, 0, 0, 0, 2, 0, 0, 0, 2, 2, 0, 0, & !
0, 1, 1, 0, 0, 1, 1, 0, 0, 2, 0, 0, 0, 2, 0, 0, 0, 2, 2, 0, & !
0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 2, 0, 0, 0, 2, 0, 0, 0, 2, 2, & !
1, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 2, 0, 0, 0, 2, 2, 0, 0, 2, & ! 25
0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 2, 2, 2, 2, 0, 0, 0, 0, & !
3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4 & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
FE_cellnodeParentnodeWeights(1:FE_Nnodes(me),1:FE_Ncellnodes(FE_geomtype(me)),me) = & ! element 21 (3D 20node 27ip)
reshape(real([&
changes towards supporting new Marc2017 input file format still not working
2018-01-10 21:43:25 +05:30
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & ! 5
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & ! 10
0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, & ! 15
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, & !
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, & !
0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, & !
0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, & !
0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, & ! 20
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, & !
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, & !
0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, & !
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, & !
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, & ! 25
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, & !
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, & ! 30
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, & !
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, & !
4, 1, 1, 1, 0, 0, 0, 0, 8, 2, 2, 8, 0, 0, 0, 0, 0, 0, 0, 0, & !
1, 4, 1, 1, 0, 0, 0, 0, 8, 8, 2, 2, 0, 0, 0, 0, 0, 0, 0, 0, & !
1, 1, 4, 1, 0, 0, 0, 0, 2, 8, 8, 2, 0, 0, 0, 0, 0, 0, 0, 0, & ! 35
1, 1, 1, 4, 0, 0, 0, 0, 2, 2, 8, 8, 0, 0, 0, 0, 0, 0, 0, 0, & !
4, 1, 0, 0, 1, 1, 0, 0, 8, 0, 0, 0, 2, 0, 0, 0, 8, 2, 0, 0, & !
1, 4, 0, 0, 1, 1, 0, 0, 8, 0, 0, 0, 2, 0, 0, 0, 2, 8, 0, 0, & !
0, 4, 1, 0, 0, 1, 1, 0, 0, 8, 0, 0, 0, 2, 0, 0, 0, 8, 2, 0, & !
0, 1, 4, 0, 0, 1, 1, 0, 0, 8, 0, 0, 0, 2, 0, 0, 0, 2, 8, 0, & ! 40
0, 0, 4, 1, 0, 0, 1, 1, 0, 0, 8, 0, 0, 0, 2, 0, 0, 0, 8, 2, & !
0, 0, 1, 4, 0, 0, 1, 1, 0, 0, 8, 0, 0, 0, 2, 0, 0, 0, 2, 8, & !
1, 0, 0, 4, 1, 0, 0, 1, 0, 0, 0, 8, 0, 0, 0, 2, 2, 0, 0, 8, & !
4, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 8, 0, 0, 0, 2, 8, 0, 0, 2, & !
1, 1, 0, 0, 4, 1, 0, 0, 2, 0, 0, 0, 8, 0, 0, 0, 8, 2, 0, 0, & ! 45
1, 1, 0, 0, 1, 4, 0, 0, 2, 0, 0, 0, 8, 0, 0, 0, 2, 8, 0, 0, & !
0, 1, 1, 0, 0, 4, 1, 0, 0, 2, 0, 0, 0, 8, 0, 0, 0, 8, 2, 0, & !
0, 1, 1, 0, 0, 1, 4, 0, 0, 2, 0, 0, 0, 8, 0, 0, 0, 2, 8, 0, & !
0, 0, 1, 1, 0, 0, 4, 1, 0, 0, 2, 0, 0, 0, 8, 0, 0, 0, 8, 2, & !
0, 0, 1, 1, 0, 0, 1, 4, 0, 0, 2, 0, 0, 0, 8, 0, 0, 0, 2, 8, & ! 50
1, 0, 0, 1, 1, 0, 0, 4, 0, 0, 0, 2, 0, 0, 0, 8, 2, 0, 0, 8, & !
1, 0, 0, 1, 4, 0, 0, 1, 0, 0, 0, 2, 0, 0, 0, 8, 8, 0, 0, 2, & !
0, 0, 0, 0, 4, 1, 1, 1, 0, 0, 0, 0, 8, 2, 2, 8, 0, 0, 0, 0, & !
0, 0, 0, 0, 1, 4, 1, 1, 0, 0, 0, 0, 8, 8, 2, 2, 0, 0, 0, 0, & !
0, 0, 0, 0, 1, 1, 4, 1, 0, 0, 0, 0, 2, 8, 8, 2, 0, 0, 0, 0, & ! 55
0, 0, 0, 0, 1, 1, 1, 4, 0, 0, 0, 0, 2, 2, 8, 8, 0, 0, 0, 0, & !
24, 8, 4, 8, 8, 4, 3, 4, 32,12,12,32, 12, 4, 4,12, 32,12, 4,12, & !
8,24, 8, 4, 4, 8, 4, 3, 32,32,12,12, 12,12, 4, 4, 12,32,12, 4, & !
4, 8,24, 8, 3, 4, 8, 4, 12,32,32,12, 4,12,12, 4, 4,12,32,12, & !
8, 4, 8,24, 4, 3, 4, 8, 12,12,32,32, 4, 4,12,12, 12, 4,12,32, & ! 60
8, 4, 3, 4, 24, 8, 4, 8, 12, 4, 4,12, 32,12,12,32, 32,12, 4,12, & !
4, 8, 4, 3, 8,24, 8, 4, 12,12, 4, 4, 32,32,12,12, 12,32,12, 4, & !
3, 4, 8, 4, 4, 8,24, 8, 4,12,12, 4, 12,32,32,12, 4,12,32,12, & !
4, 3, 4, 8, 8, 4, 8,24, 4, 4,12,12, 12,12,32,32, 12, 4,12,32 & !
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
],pReal),[FE_Nnodes(me),FE_Ncellnodes(FE_geomtype(me))])
! *** FE_cellface ***
me = 0_pInt
me = me + 1_pInt
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
FE_cellface(1:FE_NcellnodesPerCellface(me),1:FE_NipNeighbors(me),me) = & ! 2D 3node, VTK_TRIANGLE (5)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
2,3, &
3,1, &
1,2 &
],pInt),[FE_NcellnodesPerCellface(me),FE_NipNeighbors(me)])
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
FE_cellface(1:FE_NcellnodesPerCellface(me),1:FE_NipNeighbors(me),me) = & ! 2D 4node, VTK_QUAD (9)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
2,3, &
4,1, &
3,4, &
1,2 &
],pInt),[FE_NcellnodesPerCellface(me),FE_NipNeighbors(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
FE_cellface(1:FE_NcellnodesPerCellface(me),1:FE_NipNeighbors(me),me) = & ! 3D 4node, VTK_TETRA (10)
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
1,3,2, &
1,2,4, &
2,3,4, &
1,4,3 &
],pInt),[FE_NcellnodesPerCellface(me),FE_NipNeighbors(me)])
switched element library to geomType based. saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count. introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-16 04:15:20 +05:30
me = me + 1_pInt
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
FE_cellface(1:FE_NcellnodesPerCellface(me),1:FE_NipNeighbors(me),me) = & ! 3D 8node, VTK_HEXAHEDRON (12)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
reshape(int([&
Major restructuring of mesh.f90 due to introduction of "cells". A cell is the volume (area in 2D elements) that surrounds an ip. It is described by up to 8 cell nodes and is used to calculate ip volume, area, and face normals. It can also be used to generate a mesh for e.g. paraview that contains ip output as cell centered data. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! For marc simulations, run ./code/setup/setup_code.sh !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2013-04-15 13:43:20 +05:30
2,3,7,6, &
4,1,5,8, &
3,4,8,7, &
1,2,6,5, &
5,6,7,8, &
1,4,3,2 &
],pInt),[FE_NcellnodesPerCellface(me),FE_NipNeighbors(me)])
initialize element
2019-01-24 20:47:20 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end subroutine mesh_build_FEdata
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
fixed reading/writing of integer arrays with function for real arrays
2012-08-16 20:25:23 +05:30
end module mesh