programmer's documentation
 All Data Structures Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
Data setting for drift scalars for coal combustion

Introduction

This page provides an example of code blocks that may be used to perform a calculation with drift scalars for coal combustion.

Physical properties

Local variables to be added

The following local variables need to be defined for the examples in this section:

integer ivart, iel, ifac
integer ipcvis, ipccp
integer ipcvsl, ivar
integer f_id
integer iclapc, idecal
integer ll
integer ipcte1, icla, iromf
integer iscdri, keydri, iflid, nfld, keyccl
double precision xrtp
double precision aa1, bb1, cc1, dd1
double precision aa2, bb2, cc2, dd2
double precision aa3, bb3, cc3, dd3
double precision aa4, bb4, cc4, dd4
double precision aa5, bb5, cc5, dd5
double precision aa6, bb6, cc6, dd6
double precision aa7, bb7, cc7, dd7
double precision visco_o2, visco_co, visco_h2, visco_n2
double precision visco_so2, visco_nh3, visco_co2
double precision g
character*80 fname
double precision, allocatable, dimension(:) :: x1, visco
double precision, dimension(:), pointer :: taup
double precision, dimension(:), pointer :: taupg

Initialization and finalization

The following initialization block needs to be added for the following examples:

ipcvis = ipproc(iviscl)
allocate(x1(ncelet), visco(ncelet))
! Key id for drift scalar
call field_get_key_id("drift_scalar_model", keydri)
! Key id of the coal scalar class
call field_get_key_id("scalar_class", keyccl)
! Number of fields

In theory Fortran 95 deallocates locally-allocated arrays automatically, but deallocating arrays in a symmetric manner to their allocation is good practice, and it avoids using a different logic for C and Fortran.

Body

Here is the corresponding code:

g = sqrt(gx**2 + gy**2 + gz**2)
! Temperature
ipcte1 = ipproc(itemp1)
! Gas density
iromf = ipproc(irom1)
! First initialization
if (ntcabs.le.1) then
do iel = 1, ncel
visco(iel) = viscl0
propce(iel,iromf) = ro0
do icla = 1, nclacp
propce(iel,ipproc(irom2(icla))) = ro0
propce(iel,ipproc(idiam2(icla))) = diam20(icla)
enddo
enddo
endif
!----------------Gas viscosity function of temperature--------------------------
!
!--------------------------1-O2 2-CO 3-H2 4-N2 5-SO2 6-NH3 7-CO2----------------
aa1 = 4.0495d-6
bb1 = 6.22d-8
cc1 = -2.3032d-11
dd1 = 4.4077d-15
aa2 = 9.9987d-6
bb2 = 5.1578d-8
cc2 = -1.8383d-11
dd2 = 3.33307d-15
aa3 = 2.894d-6
bb3 = 2.22508d-8
cc3 = -8.041d-12
dd3 = 1.4619d-15
aa4 = 4.3093d-6
bb4 = 5.0516d-8
cc4 = -1.7869d-11
dd4 = 3.2136d-15
aa5 = -1.9889d-6
bb5 = 5.365d-8
cc5 = -1.4286d-11
dd5 = 2.1639d-15
aa6 = -1.293d-6
bb6 = 4.1194d-8
cc6 = -1.772d-11
dd6 = 1.8699d-15
aa7 = 4.4822d-7
bb7 = 5.4327d-8
cc7 = -1.7581d-11
dd7 = 2.9979d-15
!-------------------------------------------------------------------------------
! law mu = a + b T + c T**2 + d T**3
! so propce(iel, ipcvis) = a +b*xrtp+c*xrtp**2 + d*xrtp**3
!-------------------------------------------------------------------------------
if (ntcabs.gt.1) then
do iel = 1, ncel
xrtp = propce(iel,ipcte1)
visco_o2 = aa1 + xrtp*bb1 + cc1*xrtp**2 + dd1*xrtp**3
visco_co = aa2 + xrtp*bb2 + cc2*xrtp**2 + dd2*xrtp**3
visco_h2 = aa3 + xrtp*bb3 + cc3*xrtp**2 + dd3*xrtp**3
visco_n2 = aa4 + xrtp*bb4 + cc4*xrtp**2 + dd4*xrtp**3
visco_so2 = aa5 + xrtp*bb5 + cc5*xrtp**2 + dd5*xrtp**3
visco_nh3 = aa6 + xrtp*bb6 + cc6*xrtp**2 + dd6*xrtp**3
visco_co2 = aa7 + xrtp*bb7 + cc7*xrtp**2 + dd7*xrtp**3
! Viscosity of the mixing
visco(iel) = ( propce(iel,ipproc(iym1(8))) * visco_o2 &
+ propce(iel,ipproc(iym1(3))) * visco_co &
+ propce(iel,ipproc(iym1(5))) * visco_h2 &
+ propce(iel,ipproc(iym1(12)))* visco_n2 &
+ propce(iel,ipproc(iym1(11)))* visco_so2 &
+ propce(iel,ipproc(iym1(7))) * visco_nh3 &
+ propce(iel,ipproc(iym1(9))) * visco_co2 )/ &
( propce(iel,ipproc(iym1(8))) + propce(iel,ipproc(iym1(3))) &
+ propce(iel,ipproc(iym1(5))) + propce(iel,ipproc(iym1(12))) &
+ propce(iel,ipproc(iym1(11)))+ propce(iel,ipproc(iym1(7))) &
+ propce(iel,ipproc(iym1(9))))
enddo
endif
! Compute x1 = 1 - sum x2
do iel = 1, ncel
x1(iel) = 1.d0
do icla = 1, nclacp
x1(iel) = x1(iel) - propce(iel,ipproc(ix2(icla))) !FIXME is propce(iel,ipproc(ix2(icla))) initialized ?
enddo
enddo
! All gas scalars have the same drift as if1m(1)
!-----------------------------------------------
do iflid = 0, nfld-1
! Index of the scalar class (<0 if the scalar belongs to the gas phase)
call field_get_key_int(iflid, keyccl, icla)
call field_get_key_int(iflid, keydri, iscdri)
! We only handle here one scalar with a drift per gas class
if (icla.le.-1.and.btest(iscdri, drift_scalar_add_drift_flux)) then
! Position of variables, coefficients
! -----------------------------------
! Name of the drift scalar
call field_get_name(iflid, fname)
! Index of the corresponding "relaxation time" (taupg) for the gas
! WARNING: for the gas, this tau might be negative
call field_get_id('drift_tau_'//trim(fname), f_id)
call field_get_val_s(f_id, taupg)
! Initialize to 0
do iel = 1, ncel
taupg(iel) = 0.d0
enddo
endif
enddo
! Loop over coal particle classes
! We only handle here coal class with a drift
!--------------------------------------------
do iflid = 0, nfld-1
! index of the coal particle class
call field_get_key_int(iflid, keyccl, icla)
call field_get_key_int(iflid, keydri, iscdri)
! We only handle here one scalar with a drift per particle class
if (icla.ge.1.and.btest(iscdri, drift_scalar_add_drift_flux)) then
! Position of variables, coefficients
! -----------------------------------
! Name of the drift scalar
call field_get_name(iflid, fname)
! Index of the corresponding relaxation time (taup)
call field_get_id('drift_tau_'//trim(fname), f_id)
call field_get_val_s(f_id, taup)
! Computation of the relaxation time of the particles
! the drift is therefore v_g = tau_p * g
!----------------------------------------------------
do iel = 1, ncel
! Simple model for Low Reynolds Numbers
taup(iel) = x1(iel) * propce(iel,ipproc(irom2(icla))) &
* propce(iel,ipproc(idiam2(icla)))**2 &
/ (18.d0*visco(iel))
enddo
! Drift for the gas:
! tau_pg = - Sum_i X2_i v_gi
do iel = 1, ncel
taupg(iel) = taupg(iel) &
- ( taup(iel) * propce(iel,ipproc(ix2(icla))) )
enddo
endif
enddo