!------------------------------------------------------------------------------- ! Code_Saturne version 2.0.0-rc1 ! -------------------------- ! This file is part of the Code_Saturne Kernel, element of the ! Code_Saturne CFD tool. ! Copyright (C) 1998-2009 EDF S.A., France ! contact: saturne-support@edf.fr ! The Code_Saturne Kernel is free software; you can redistribute it ! and/or modify it under the terms of the GNU General Public License ! as published by the Free Software Foundation; either version 2 of ! the License, or (at your option) any later version. ! The Code_Saturne Kernel is distributed in the hope that it will be ! useful, but WITHOUT ANY WARRANTY; without even the implied warranty ! of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ! GNU General Public License for more details. ! You should have received a copy of the GNU General Public License ! along with the Code_Saturne Kernel; if not, write to the ! Free Software Foundation, Inc., ! 51 Franklin St, Fifth Floor, ! Boston, MA 02110-1301 USA !------------------------------------------------------------------------------- subroutine usvpst & !================ ( idbia0 , idbra0 , ipart , & ndim , ncelet , ncel , nfac , nfabor , nfml , nprfml , & nnod , lndfac , lndfbr , ncelbr , & nvar , nscal , nphas , nvlsta , & ncelps , nfacps , nfbrps , & nideve , nrdeve , nituse , nrtuse , & itypps , ifacel , ifabor , ifmfbr , ifmcel , iprfml , & ipnfac , nodfac , ipnfbr , nodfbr , & lstcel , lstfac , lstfbr , & idevel , ituser , ia , & xyzcen , surfac , surfbo , cdgfac , cdgfbo , xyznod , volume , & dt , rtpa , rtp , propce , propfa , propfb , & coefa , coefb , statis , & tracel , trafac , trafbr , rdevel , rtuser , ra ) !=============================================================================== ! Purpose: ! ------- ! User subroutine. ! Output additional variables on a postprocessing mesh. ! Several "automatic" postprocessing meshes may be defined: ! - The volume mesh (ipart=-1) if 'ichrvl' = 1 ! - The boundary mesh (ipart=-2) if 'ichrbo' = 1 ! - SYRTHES coupling surface (ipart < -2) if 'ichrsy' = 1 ! - Cooling tower exchange zone meshes (ipart < -2) if 'ichrze' = 1 ! ! Additional meshes (cells or faces) may also be defined using the ! 'usdpst' user subroutine, (and possibly modified using 'usmpst'). ! This subroutine is called once for each post-processing mesh ! (with a different value of 'ipart') for each time step at which output ! on this mesh is active. !------------------------------------------------------------------------------- ! Arguments !__________________.____._____.________________________________________________. ! name !type!mode ! role ! !__________________!____!_____!________________________________________________! ! idbia0 ! i ! <-- ! number of first free position in ia ! ! idbra0 ! i ! <-- ! number of first free position in ra ! ! ipart ! i ! <-- ! number of the post-processing mesh (< 0 or > 0)! ! ndim ! i ! <-- ! spatial dimension ! ! ncelet ! i ! <-- ! number of extended (real + ghost) cells ! ! ncel ! i ! <-- ! number of cells ! ! nfac ! i ! <-- ! number of interior faces ! ! nfabor ! i ! <-- ! number of boundary faces ! ! nfml ! i ! <-- ! number of families (group classes) ! ! nprfml ! i ! <-- ! number of properties per family (group class) ! ! nnod ! i ! <-- ! number of vertices ! ! lndfac ! i ! <-- ! size of nodfac indexed array ! ! lndfbr ! i ! <-- ! size of nodfbr indexed array ! ! ncelbr ! i ! <-- ! number of cells with faces on boundary ! ! nvar ! i ! <-- ! total number of variables ! ! nscal ! i ! <-- ! total number of scalars ! ! nphas ! i ! <-- ! number of phases ! ! nvlsta ! i ! <-- ! number of Lagrangian statistical variables ! ! ncelps ! i ! <-- ! number of cells in post-processing mesh ! ! nfacps ! i ! <-- ! number of interior faces in post-process. mesh ! ! nfbrps ! i ! <-- ! number of boundary faces in post-process. mesh ! ! nideve, nrdeve ! i ! <-- ! sizes of idevel and rdevel arrays ! ! nituse, nrtuse ! i ! <-- ! sizes of ituser and rtuser arrays ! ! itypps(3) ! ia ! <-- ! global presence flag (0 or 1) for cells (1), ! ! ! ! ! interior faces (2), or boundary faces (3) in ! ! ! ! ! post-processing mesh ! ! ifacel(2, nfac) ! ia ! <-- ! interior faces -> cells connectivity ! ! ifabor(nfabor) ! ia ! <-- ! boundary faces -> cells connectivity ! ! ifmfbr(nfabor) ! ia ! <-- ! boundary face family numbers ! ! ifmcel(ncelet) ! ia ! <-- ! cell family numbers ! ! iprfml ! ia ! <-- ! property numbers per family ! ! (nfml, nprfml) ! ! ! ! ! ipnfac(nfac+1) ! ia ! <-- ! interior faces -> vertices index (optional) ! ! nodfac(lndfac) ! ia ! <-- ! interior faces -> vertices list (optional) ! ! ipnfbr(nfabor+1) ! ia ! <-- ! boundary faces -> vertices index (optional) ! ! nodfbr(lndfbr) ! ia ! <-- ! boundary faces -> vertices list (optional) ! ! lstcel(ncelps) ! ia ! <-- ! list of cells in post-processing mesh ! ! lstfac(nfacps) ! ia ! <-- ! list of interior faces in post-processing mesh ! ! lstfbr(nfbrps) ! ia ! <-- ! list of boundary faces in post-processing mesh ! ! idevel(nideve) ! ia ! <-> ! integer work array for temporary development ! ! ituser(nituse) ! ia ! <-> ! user-reserved integer work array ! ! ia(*) ! ia ! --- ! main integer work array ! ! xyzcen ! ra ! <-- ! cell centers ! ! (ndim, ncelet) ! ! ! ! ! surfac ! ra ! <-- ! interior faces surface vectors ! ! (ndim, nfac) ! ! ! ! ! surfbo ! ra ! <-- ! boundary faces surface vectors ! ! (ndim, nfabor) ! ! ! ! ! cdgfac ! ra ! <-- ! interior faces centers of gravity ! ! (ndim, nfac) ! ! ! ! ! cdgfbo ! ra ! <-- ! boundary faces centers of gravity ! ! (ndim, nfabor) ! ! ! ! ! xyznod ! ra ! <-- ! vertex coordinates (optional) ! ! (ndim, nnod) ! ! ! ! ! volume(ncelet) ! ra ! <-- ! cell volumes ! ! dt(ncelet) ! ra ! <-- ! time step (per cell) ! ! rtp, rtpa ! ra ! <-- ! calculated variables at cell centers ! ! (ncelet, *) ! ! ! (at current and previous time steps) ! ! propce(ncelet, *)! ra ! <-- ! physical properties at cell centers ! ! propfa(nfac, *) ! ra ! <-- ! physical properties at interior face centers ! ! propfb(nfabor, *)! ra ! <-- ! physical properties at boundary face centers ! ! coefa, coefb ! ra ! <-- ! boundary conditions ! ! (nfabor, *) ! ! ! ! ! statis ! ra ! <-- ! statistic values (Lagrangian) ! ! (ncelet, nvlsta)! ! ! ! ! tracel(*) ! ra ! --- ! work array for post-processed cell values ! ! trafac(*) ! ra ! --- ! work array for post-processed face values ! ! trafbr(*) ! ra ! --- ! work array for post-processed boundary face v. ! ! rdevel(nrdeve) ! ra ! <-> ! real work array for temporary development ! ! rtuser(nrtuse) ! ra ! <-> ! user-reserved real work array ! ! ra(*) ! ra ! --- ! main real work array ! !__________________!____!_____!________________________________________________! ! Type: i (integer), r (real), s (string), a (array), l (logical), ! and composite types (ex: ra real array) ! mode: <-- input, --> output, <-> modifies data, --- work array !=============================================================================== implicit none !=============================================================================== ! Common blocks !=============================================================================== include "paramx.h" include "cstnum.h" include "pointe.h" include "entsor.h" include "optcal.h" include "numvar.h" include "parall.h" include "period.h" !=============================================================================== ! Arguments integer idbia0, idbra0 integer ipart integer ndim, ncelet, ncel, nfac, nfabor integer nfml, nprfml integer nnod, lndfac, lndfbr, ncelbr integer nvar, nscal , nphas , nvlsta integer ncelps, nfacps, nfbrps integer nideve, nrdeve, nituse, nrtuse integer itypps(3) integer ifacel(2,nfac), ifabor(nfabor) integer ifmfbr(nfabor), ifmcel(ncelet) integer iprfml(nfml,nprfml) integer ipnfac(nfac+1), nodfac(lndfac) integer ipnfbr(nfabor+1), nodfbr(lndfbr) integer lstcel(ncelps), lstfac(nfacps), lstfbr(nfbrps) integer idevel(nideve), ituser(nituse), ia(*) double precision xyzcen(ndim,ncelet) double precision surfac(ndim,nfac), surfbo(ndim,nfabor) double precision cdgfac(ndim,nfac), cdgfbo(ndim,nfabor) double precision xyznod(ndim,nnod), volume(ncelet) double precision dt(ncelet), rtpa(ncelet,*), rtp(ncelet,*) double precision propce(ncelet,*) double precision propfa(nfac,*), propfb(nfabor,*) double precision coefa(nfabor,*), coefb(nfabor,*) double precision statis(ncelet,nvlsta) double precision tracel(ncelps*3) double precision trafac(nfacps*3), trafbr(nfbrps*3) double precision rdevel(nrdeve), rtuser(nrtuse) double precision ra(*) ! Local variables character*32 namevr integer ntindp integer iel, ifac, iloc, iphas, ivar, iclt integer idimt, ii , jj integer idimte, itenso, ientla, ivarpr integer imom1, imom2, ipcmo1, ipcmo2, idtcm double precision pond double precision rvoid(1) integer ipass data ipass /0/ save ipass !=============================================================================== !=============================================================================== ! 1. Handle variables to output ! MUST BE FILLED IN by the user at indicated places !=============================================================================== ! A post-processing id a "part" (using the EnSight vocabulary; the MED and ! CGNS equivalents are "mesh" and "base" respectively). ! The user will have defined post-processing meshes in 'usdpst' ('nbpart' ! post-processing meshes). ! This subroutine is called once for each post-processing mesh ! (with a different value of 'ipart') for each time step at which output ! on this mesh is active. For each mesh and for all variables we wish to ! post-process here, we must define certain parameters and pass them to ! the 'psteva' subroutine, which is in charge of the actual output. ! These parameters are: ! namevr <-- variable name ! idimt <-- variable dimension ! (1: scalar, 3: vector, 6: symmetric tensor, 9: tensor) ! ientla <-- when idimt >1, this flag specifies if the array containing the ! variable values is interlaced when ientla = 1 ! (x1, y1, z1, x2, y2, z2, x3, y3, z3...), or non-interlaced ! when ientla = 0 (x1,x2,x3,...,y1,y2,y3,...,z1,z2,z3,...). ! ivarpr <-- specifies if the array containing the variable is defined on ! the "parent" mesh or locally. ! Even if the 'ipart' post-processing mesh contains all the ! elements of its parent mesh, their numbering may be different, ! especially when different element types are present. ! The 'tracel' array passed as an argument to 'psteva' is built ! relative to the numbering of the 'ipart' post-processing mesh. ! To post-process a variable contained for example in the 'rtuser' ! array, it should first be re-ordered, as shown here: ! do iloc = 1, ncelps ! iel = lstcel(iloc) ! tracel(iloc) = rtuser(iel) ! enddo ! An alternative option is provided, to avoid unnecessary copies: ! an array defined on the parent mesh, such our 'rtuser' example, ! may be passed directly to 'psteva', specifying that values ! are defined on the parent mesh instead of the post-processing mesh, ! by setting the 'ivarpr' argument of 'psteva' to 1. ! Note: be cautious with variable name lengths. ! We allow up to 32 characters here, but names may be truncted depending on the ! output format: ! - 19 characters for EnSight ! - 32 characters for MED ! The nam length is not limited internally, so in case of 2 variables whoses ! names differ only after the 19th character, the corresponding names will ! both appear in the ".case" file; simply renaming one of the field descriptors ! in this text file will correct the output. ! Whitespace at the beginning or the end of a line is truncated automatically. ! Depending on the format used, prohibited characters (under EnSight, characters ! ( ) ] [ + - @ ! # * ^ $ / as well as white spaces and tabulations ! are automatically replaced by the _ character. ! Examples: ! For post-processing mesh 2, we output the velocity, pressure, and prescribed ! temperature at boundary faces (as well as 0 on possible interior faces) ! For post-processing mesh 1, we output all the variables usually ! post-processed, using a more compact coding. ! Examples given here correspond to the meshes defined in usdpst.f90 and ! modified in usmpst.f90. !=============================================================================== ! 1.1. Examples of volume variables on the main volume mesh (ipart = -1) !=============================================================================== if (ipart .eq. -1) then ! 1.1.1 Output of k=1/2(R11+R22+R33) for the Rij-epsilon model ! ------------------------------------------------------ iphas = 1 ! We only consider phase 1 in this example if (itytur(iphas) .eq. 3) then ! Initialize variable name do ii = 1, 32 namevr(ii:ii) = ' ' enddo ! Variable name namevr = 'Turb energy' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 do iloc = 1, ncelps iel = lstcel(iloc) tracel(iloc) = 0.5d0*( rtp(iel,ir11(iphas)) & + rtp(iel,ir22(iphas)) & + rtp(iel,ir33(iphas)) ) enddo ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values; as we have no face values, we can pass a ! trivial array rvoid instead of trafac and trafbr. call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, rvoid, rvoid) endif ! 1.1.2 Output of a combination of moments ! ---------------------------------- ! We assume in this example that we have 2 temporal means (moments): ! for imom=1 ! for imom=2 ! We seek to plot =-**2 iphas = 1 ! We only consider phase 1 in this example if (nbmomt .ge. 2) then ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Moment numbers: imom1 = 1 imom2 = 2 ! Position in 'propce' of the array of temporal accumulation for moments, ! propce(iel,ipcmom) ipcmo1 = ipproc(icmome(imom1)) ipcmo2 = ipproc(icmome(imom2)) ! Variable name namevr = '' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 ! The temporal accumulation for moments must be divided by the accumulated ! time, which id an array of size ncel or a single real number: ! - array of size ncel if idtmom(imom) > 0 : propce(iel, idtcm) ! - or simple real if idtmom(imom) < 0 : dtcmom(idtcm) ! To improve this example's readability, we assume moments imom1 and imom2 ! have been computed on the same time window. if(idtmom(imom1).gt.0) then idtcm = ipproc(icdtmo(idtmom(imom1))) do iloc = 1, ncelps iel = lstcel(iloc) tracel(iloc) = propce(iel,ipcmo2)/max(propce(iel,idtcm),epzero) & - (propce(iel,ipcmo1)/max(propce(iel,idtcm),epzero))**2 enddo elseif(idtmom(imom1).lt.0) then idtcm = -idtmom(imom1) do iloc = 1, ncelps iel = lstcel(iloc) tracel(iloc) = propce(iel,ipcmo2)/max(dtcmom(idtcm),epzero) & - (propce(iel,ipcmo1)/max(dtcmom(idtcm),epzero))**2 enddo endif ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values; as we have no face values, we can pass a ! trivial array rvoid instead of trafac and trafbr. call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, rvoid, rvoid) endif !=============================================================================== ! 1.2. Examples of volume variables on the boundary mesh (ipart = -2) !=============================================================================== else if (ipart .eq. -2) then ! 1.2.1 Output of the density at the boundary ! ------------------------------------- iphas = 1 ! We only consider phase 1 in this example ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'Density at boundary' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 ! We directly use the propfb array defined on the parent mesh. ivarpr = 1 ! Output values; as we have only boundary face values, we can pass a ! trivial array rvoid instead of tracel and trafac. call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, rvoid, rvoid, & propfb(1,ipprob(irom(iphas)))) ! 1.2.2 Output of the domain number in parallel ! --------------------------------------- ! This variable is independent of time, so we output it once ! only (see 'ntindp' below) if (ipass.eq.0 .and. irangp.ge.0) then ipass = ipass + 1 ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'domain number' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 do iloc = 1, nfbrps ! ifac = lstfbr(iloc) trafbr(iloc) = irangp + 1 enddo ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 ! Values are defined on the work array, not on the parent. ivarpr = 0 ! This variable is time-invariant; ! We assign a negative time to it and output it once only to avoid ! duplicating it at each output time. ntindp = -1 ! Output values; as we have only boundary face values, we can pass a ! trivial array rvoid instead of trafac and trafbr. call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntindp, ttcabs, rvoid, rvoid, trafbr) endif !=============================================================================== ! 1.3. Examples of volume variables on user meshes 1 or 2 !=============================================================================== else if (ipart.eq.1 .or. ipart.eq.2) then ! 1.3.1 Output of the velocity ! ---------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'Interpol velocity' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 3 ! Values are interlaced. ientla = 1 iphas = 1 ! We only consider phase 1 in this example ! Compute variable values on interior faces. ! In this example, we use a simple linear interpolation. ! For parallel calculations, if neighbors are used, they must be synchronized ! first. This also applies for periodicity. When both are used, parallel ! synchronization must always be done before periodic synchronization. if(irangp.ge.0) then call parcom (rtp(1,iu(iphas))) call parcom (rtp(1,iv(iphas))) call parcom (rtp(1,iw(iphas))) endif if(iperio.eq.1) then idimte = 1 itenso = 0 call percom(idimte, itenso, & !========== rtp(1,iu(iphas)), rtp(1,iu(iphas)), rtp(1,iu(iphas)), & rtp(1,iv(iphas)), rtp(1,iv(iphas)), rtp(1,iv(iphas)), & rtp(1,iw(iphas)), rtp(1,iw(iphas)), rtp(1,iw(iphas))) endif do iloc = 1, nfacps ifac = lstfac(iloc) ii = ifacel(1, ifac) jj = ifacel(2, ifac) pond = ra(ipond-1+ifac) trafac(1 + (iloc-1)*idimt) = pond * rtp(ii,iu(iphas)) & + (1.d0 - pond) * rtp(jj,iu(iphas)) trafac(2 + (iloc-1)*idimt) = pond * rtp(ii,iv(iphas)) & + (1.d0 - pond) * rtp(jj,iv(iphas)) trafac(3 + (iloc-1)*idimt) = pond * rtp(ii,iw(iphas)) & + (1.d0 - pond) * rtp(jj,iw(iphas)) enddo ! Compute variable values on boundary faces. ! In this example, we use a simple copy of the adjacent cell value. do iloc = 1, nfbrps ifac = lstfbr(iloc) ii = ifabor(ifac) trafbr(1 + (iloc-1)*idimt) = rtp(ii, iu(iphas)) trafbr(2 + (iloc-1)*idimt) = rtp(ii, iv(iphas)) trafbr(3 + (iloc-1)*idimt) = rtp(ii, iw(iphas)) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, trafac, trafbr) ! 1.3.2 Output of the pressure ! ---------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'Interpol pressure' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 iphas = 1 ! We only consider phase 1 in this example ! Variable number ivar = ipr(iphas) ! Compute variable values on interior faces. ! In this example, we use a simple linear interpolation. ! For parallel calculations, if neighbors are used, they must be synchronized ! first. This also applies for periodicity. When both are used, parallel ! synchronization must always be done before periodic synchronization. if(irangp.ge.0) then call parcom (rtp(1,ivar)) endif if(iperio.eq.1) then idimte = 0 itenso = 0 call percom(idimte, itenso, & !========== rtp(1,ivar), rtp(1,ivar), rtp(1,ivar), & rtp(1,ivar), rtp(1,ivar), rtp(1,ivar), & rtp(1,ivar), rtp(1,ivar), rtp(1,ivar)) endif do iloc = 1, nfacps ifac = lstfac(iloc) ii = ifacel(1, ifac) jj = ifacel(2, ifac) pond = ra(ipond-1+ifac) trafac(iloc) = pond * rtp(ii, ivar) & + (1.d0 - pond) * rtp(jj, ivar) enddo ! Compute variable values on boundary faces. ! In this example, we use a simple copy of the adjacent cell value. do iloc = 1, nfbrps ifac = lstfbr(iloc) ii = ifabor(ifac) trafbr(iloc) = rtp(ii, ivar) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, trafac, trafbr) ! 1.3.3 Output of the boundary temperature ! ---------------------------------- iphas = 1 ! We only consider phase 1 in this example if (iscalt(iphas) .gt. 0) then ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'Boundary temperature' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 ! Set value to 0 for interior faces do iloc = 1, nfacps trafac(iloc) = 0.d0 enddo ! Compute variable values on boundary faces. ivar = isca(iscalt(iphas)) iclt = iclrtp(ivar,icoef) do iloc = 1, nfbrps ifac = lstfbr(iloc) ii = ifabor(ifac) trafbr(iloc) = coefa(ifac,iclt)+coefb(ifac,iclt)*rtp(ii, ivar) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, trafac, trafbr) endif ! The examples below illustrate how to output a same variable in different ! ways (interlaced or not, using an indirection or not). ! 1.3.4 Output of the centers of gravity, interlaced ! -------------------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'face cog (interlaced)' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 3 ! Values are interlaced ientla = 1 do iloc = 1, nfacps ifac = lstfac(iloc) trafac(1 + (iloc-1)*idimt ) = cdgfac(1, ifac) trafac(2 + (iloc-1)*idimt ) = cdgfac(2, ifac) trafac(3 + (iloc-1)*idimt ) = cdgfac(3, ifac) enddo ! Compute variable values on boundary faces do iloc = 1, nfbrps ifac = lstfbr(iloc) trafbr(1 + (iloc-1)*idimt ) = cdgfbo(1, ifac) trafbr(2 + (iloc-1)*idimt ) = cdgfbo(2, ifac) trafbr(3 + (iloc-1)*idimt ) = cdgfbo(3, ifac) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, trafac, trafbr) ! 1.3.5 Output of the centers of gravity, non-interlaced ! -------------------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'face cog (non-interlaced)' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 3 ! Values are not interlaced ientla = 0 do iloc = 1, nfacps ifac = lstfac(iloc) trafac(iloc) = cdgfac(1, ifac) trafac(iloc + nfacps) = cdgfac(2, ifac) trafac(iloc + 2*nfacps) = cdgfac(3, ifac) enddo ! Compute variable values on boundary faces do iloc = 1, nfbrps ifac = lstfbr(iloc) trafbr(iloc) = cdgfbo(1, ifac) trafbr(iloc + nfbrps) = cdgfbo(2, ifac) trafbr(iloc + 2*nfbrps) = cdgfbo(3, ifac) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, tracel, trafac, trafbr) ! 1.3.6 Output of the centers of gravity, with indirection (parent-based) ! -------------------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'face cog (parent)' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 3 ! Values are interlaced ientla = 1 ! Values are defined on the parent. ivarpr = 1 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, rvoid, cdgfac, cdgfbo) !=============================================================================== ! 1.4. Examples of volume variables on user meshes 3 or 4 !=============================================================================== else if (ipart.ge.3 .and. ipart.le.4) then ! 1.4.1 Output of the velocity ! ---------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'Velocity' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 3 ! Values are not interlaced. ientla = 0 iphas = 1 ! We only consider phase 1 in this example ! Variable number ivar = iu(iphas) ! Compute variable values on interior faces. ! In this example, we use a simple linear interpolation. ! For parallel calculations, if neighbors are used, they must be synchronized ! first. This also applies for periodicity. When both are used, parallel ! synchronization must always be done before periodic synchronization. if(irangp.ge.0) then call parcom (rtp(1,iu(iphas))) call parcom (rtp(1,iv(iphas))) call parcom (rtp(1,iw(iphas))) endif if(iperio.eq.1) then idimte = 1 itenso = 0 call percom(idimte, itenso, & !========== rtp(1,iu(iphas)), rtp(1,iu(iphas)), rtp(1,iu(iphas)), & rtp(1,iv(iphas)), rtp(1,iv(iphas)), rtp(1,iv(iphas)), & rtp(1,iw(iphas)), rtp(1,iw(iphas)), rtp(1,iw(iphas))) endif do iloc = 1, nfacps ifac = lstfac(iloc) ii = ifacel(1, ifac) jj = ifacel(2, ifac) pond = ra(ipond-1+ifac) trafac(iloc) = pond * rtp(ii, iu(iphas)) & + (1.d0 - pond) * rtp(jj, iu(iphas)) trafac(iloc + nfacps) = pond * rtp(ii, iv(iphas)) & + (1.d0 - pond) * rtp(jj, iv(iphas)) trafac(iloc + 2*nfacps) = pond * rtp(ii, iw(iphas)) & + (1.d0 - pond) * rtp(jj, iw(iphas)) enddo ! Compute variable values on boundary faces. ! In this example, we use a simple copy of the adjacent cell value. do iloc = 1, nfbrps ifac = lstfbr(iloc) ii = ifabor(ifac) trafbr(iloc ) = rtp(ii, iu(iphas)) trafbr(iloc + nfbrps) = rtp(ii, iv(iphas)) trafbr(iloc + 2*nfbrps) = rtp(ii, iw(iphas)) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, rtp(1,ivar), trafac, trafbr) ! 1.5.2 Output of the pressure ! ---------------------- ! Initialize variable name do ii = 1, 32 namevr (ii:ii) = ' ' enddo ! Variable name namevr = 'Pressure' ! Variable dimension (1: scalar, 3: vector, 6/9: symm/non-symm tensor) idimt = 1 ! Values are not interlaced (dimension 1 here, so no effect). ientla = 0 iphas = 1 ! We only consider phase 1 in this example ! Variable number ivar = ipr(iphas) ! Compute variable values on interior faces. ! In this example, we use a simple linear interpolation. ! For parallel calculations, if neighbors are used, they must be synchronized ! first. This also applies for periodicity. When both are used, parallel ! synchronization must always be done before periodic synchronization. do iloc = 1, nfacps ifac = lstfac(iloc) ii = ifacel(1, ifac) jj = ifacel(2, ifac) pond = ra(ipond-1+ifac) trafac(iloc) = pond * rtp(ii, ivar) & + (1.d0 - pond) * rtp(jj, ivar) enddo ! Compute variable values on boundary faces. ! In this example, we use a simple copy of the adjacent cell value. do iloc = 1, nfbrps ifac = lstfbr(iloc) ii = ifabor(ifac) trafbr(iloc) = rtp(ii, ivar) enddo ! Values are defined on the work array, not on the parent. ivarpr = 0 ! Output values call psteva(ipart, namevr, idimt, ientla, ivarpr, & !========== ntcabs, ttcabs, rtp(1,ivar), trafac, trafbr) endif ! end of test on post-processing mesh number return end subroutine