Professional Documents
Culture Documents
11509318
11509318
a:
LLJ
March 1979
H * fi •?• * 0f ^ Fir
Japan Atomic Energy Research Institute
JAERI l^tf- I-
««'41115'li^iU:
•SUJi Iff <U (•? ft
•'If
'I' il
•>H- Fn??
i'l'liK ;lM|i|
'?!'|ll j.'f
M-11
lt-J; \i'ii J" 'Ji'.;^;:fft1 •'• -
JAERI Report
Inquiries about the availability of reports and their reproduction should be addressed to the
Division of Technical Information, Japan Atomic Energy Research Institute, Tokai-mura, Naka-
gun, Ibaraki-ken, Japan.
S3S$*3Sfr n * IK
£ ti:
JAEKI 12fiO
Division of JPDR
A FORTRAN computer program A1K" is dcvi-lomtl for I he l'ACOM 2:10 7." romimtor
In In- capnbk' of solving oiRiln\'alm" proljlm..-. <>i' neutron (lifftision riiualimi in ciiu1. l\vo and
Ihroo spatial dinicnsions. Tin- available ccxmlinate systems are orllniKnnal i.\>, (N.Y1.
( X . Y . Z 1 and cylindrical i R . Z ) . i R . H l , (R.H./.l. T h e outer iKiundan,- condition for the
neutron flux can be chosen to be symmetric, zero flux or log derivative condition. The
present program can be used also for obtaining' the adjoint flux.
1,1,2,
2,33 ikivmr-Hm^-
次元中性子拡散コード、: ADC
Y • ADC
秋1
じi F
E、へ j!~ii,手自主 F-
1~'hj( r)
)
!i
l
¥党内i 'jiðti 研究 1 司í~ilJ !),試験か m~
I!
17
11frl J
J
!JJ
197«'|-<) 441
I"1
1'!
1
ADD
A CC . !i
4 ' 'を il'lfO
f'f fI
li FF
! AACC
OO MM2;!
;:
)W
0-
-7
755 JJ
i
Ji Jとi L LC
C/
|l
)1
aJ
J
(5 LLfた
5; c. .
l:i
i Jll'可能なJ
. Wni'liBtiMl^lt-. H
;i系I ,
.
i i(ik\¥.K'. 1X
iI'i:交!十I'.~'. .
L )
X .). ¥ X .
i.X,
'
Y1 , ((X
))
. X, , YY. , ZZ) feiU1, Im
) および. 'i
li
f il
M1i S ((R
f I'.\~町 R. Z Z)
)., ((R
R,, 0 ), ((R
ff), R. , 9O
. Z)
, Z ) -で?子#}. ?る '
> . MM'1
'
1'!. r
1 I ''
^1
" - i'
i '二 l
r る丹市川県l 条例と L て l士、対利;条f'I:.本書 'I'~I: rA条 (
1または,t
' .f-l 微分 1dl のい「れかをよ.í~ ぶこと
がで竺る でj
こ の ブ ロ グ ラ J、 i
i
l
i "I
l'
l 1
'
i'r
'kιま t
.1
:,;¥
Iできる.
骨現在,安全解析部
* as.
JAKKI 12(iO "
CONTENTS
1. Introduction 1
2. Miiin Feature of ADC Code 1
3. Details of Hie Method 2
3.1 Formulation of difference equations '.\
3. 2 Houndary conditions I
3.3 Iterative procedure for finite difference diffusion equation 5
3.4 Adjoint flux problem 11
4. Program Usage 12
4.1 Variable common and program flow path control 12
4.2 Standard program flow and required core memory l.'i
4. 3 I Mai led inpul speci f icat ion 11
4. 4 Output information 21
Acknowledgement • • 22
References 22
Appendix: Sample Input and O u t p u t "<M
.lAI-lKl 13i(l
1. n- ,m i
2. A D C ; < - l-^'M-a 1
3. J-UH'i 2
3.1. V.'it.HW.Ji.nmW '•>
3.2. JfiW-Hr- i
3.3. fiBm^WftJyP.'j^jT'^'iU': 3
3.4. Bfifl-'H'l i-<\imW& H
v u /••;.', f-tnJU; VI
4.1. i i r & i w r f i i W i . t t f v •> y ; ' . ' . • v »-.<o;i,i|jaii 12
4.2. tT.'V=n'j•/.. >/; A • v ,. -. i-ft^,,!-^!^^:, 1 ,. K!
4.3. K U f - !' 1!
4.4. l\\ ))••?-•!> 21
,.'lt ,W 22
tlM 22
lll^ 21
1. Introduction
A oimpuH'i1 program lor solving the niulli group neutron dillusinti equation is one (if thi- widely
used programs fur reactor plrysics and reactor design calculations. With I In- development ol large scale
computers, complex c.-ili-iiliit itms of neutron flux dislriluiliou in ;i reactor core such ;is burn up calculations
have become feasible l)y performing directly three dimensional calculati'ins.
Tin1 three dimensional diffusion codes, sue!) as CITATION ;ind :U)H have been developed in ..• \v;iy
I halt tlu1 programming depends on the inherent function of the digital emputer system to be applied.
The CITATION code requires large f;ist core memories for (lie calculation of a large se;de 11WR core.
On the otlier hand, the 'M)\\ code requires less fnst <-ore n)emories than the CITATION', but it requires
lotiKer coinpulion time on (lie I-'ACOM lilltl 75 because of usinn scratch disk units many times for
iteratirm processes.
The general dimensional diffusion equation program AI)C for KACOM 2'M) 75 is therefore developed
for resolving the above drawbacks. This program permits an effective use of the available core
memory by usitiK t w sli'|> I''OKTKAN job. At the first step, the size of common area and Ihe program
flow paths are arranged to be adequate for the problem to Ixi solved. These functions are called variable
common arrangement and flow path control, At Ihe nexl .step, the main program and following sub
programs are compiled according lo the generated program flow palhs. If (he storage ntjuirixl by the
problem to be solved exceeds the size of available fast memory, scratch units are used.
The finiU difference diffusion equation as well as ils adjoint equation is solved by either of the
following two iterative methods. One is a double hx>p iteration method with inner iteration and
outer iteration, where the difference equation of each energy group is solved by inner iteration usinK
ixiintwi.se or linewisc relaxation method. After the inner iteration is converged, the eigenvalue is
obtained by outer iteration. The other is a one-through iteration method which is similar to EQUIPOISK
meth'Kl. For achieving a rapid convergence of the above iteration process, the initial guess flux can be
constructed from a synthesis in our program.
The program ADC solves the eigenvalue problems of neutron diffusion equation and ils adjoint
equation in one-, two , or three-dimensions. Either symmetric, zero flux or log derivative condition is
chosen at outside boundaries. Internal control rod boundary conditions are also permitted to be used.
Various options are available to print out the neutron flux distribution averaged over the specified
regions. This computer code is programed in FORTRAN IV for the KACOM 230-75 computer.
The main features of the general dimensional diffusion calculation axle ADC are summarized as
follows;
1) Computer : The code is programmed lo be effectively used on the KACOM 2.W-75 computer.
However it can easily be made usable on any other computers having at least 61K word storage.
2) Equation to be solved : The multi-group neutron diffusion problem may be solved to obtain the
eigenvalue in one-, two- or three-dimensional geometry using finite difference formulation of diffusion
theory.
3) Geometry : Rectangular (X). (X.Y). (X.Y.Z) or cylindrical (r). (r.z), (r,0), (r,0,z).
2 A (MIIII.II I ! • • • » !¥-•• .ri.<l \ . I I I M J i J j 1 1 1 1 i . ,11 r . i h n l n i M i i (.>li \ | l l l A k k i Ilinll
3. lM;ii!s of the
The mulli group neutron diffusion equation is approximated to obtain a set of finite-difference
equations expressing neutron balance over a finite volume element. The diffusion approximation to
neutron transport equation at a location, r, and in a discrete energy group, g. is expressed by the
following differential equation 0 ) , in which a perpendicular buckling term may be allowed if required.
Tin1 fission source neutron distribution function, 1, is assumed to be only energy dependent, and the
slowing-clown source is assumed to come from only the next higher energy group.
where
To obtain t h e spatial difference equations, t h e spatial mesh points are assigned in t h e center of
three dimensional mesh volumes, or fewer dimensions, t h e mesh interval in t h e undefined dimen
sions is treated as infinite and the associated terms to t h e undefined dimensions cancel out of the
difference equations.
Equation (1) is integrated over a single mesh volume. }:o- the CI.J.K) mesh ]?>int at the position
A'- A',. > ' - Y, and Z~ Zt, t h e A' integration is performed from A',— ' ' to X, + ' ', t h e Y integration
AY AY Ay • A^
from Yj— ' ' to Y, 4 ' nnd t h e Z integration from Y.k — '}* to /-k '• / " ' . F i g u r e 1 presents a
three dimensional sketch showing t h e location of mesh point l I . J . K ) and t h e surrounding six locations
in t h e slab geometry.
The leakage t e r m s are modi lied first by transforming the volume integral of t h e l.aplacian lo a
surface integral by using t h e (Ireen's theorem:
t 2
w h e r e / J , . j , 4 . a n d D,, t,, , a r e t h e d i f f u s i o n c o e f f i c i e n t s at CI.J.K) a n d ( I , . I , K — I r r e s p e c t i v e l y , [elim-
i n a t i n g (4, f r o m t h e a b o v e equation.-;, w e g e t
' ( • 1. I - 1
The destruction and production terms of Eq. ( I ) are obtained by performing simply the integration over
the mesh volume. The destruction term is given by
2i, a, u j . i-^i. I, t$(, s, t • (6)
The production (source) term is given by
St,t,j.tV,,j.* , (7)
where
A -i"itil Ni-iunin I i Ciloil.in.in <••«!.• UN J.-M-.KI l a i n
Thus, Hie volume integration of equal inn (1 i arround mesh |«>inl I) i s v I'm 1 i leads to ihc expression
(8) where, for simplicity, the group indices are omitted.
/
.6
A
/
AZi
/ / 2 I
.3
- flXl -
'ID
where
For the zero gradient boundary condition, the associated constant, C,,;, t.. is set to zero. Kor the
logarithmic derivative boundary condition at an external or internal boundary, the tlux slope within the
finite difference volume element is assumed to be constant. The boundary condition to be satisfied at
the element surface is therefore
- ^ H l , = C, , (10)
where C, is a specified constant .
Let <fi, be the internal flux. {5, be the boundary flux and A be the interval from the internal point
to •he boundary. A linear approximation of flux within the element gives
"dxl-^'lr • or
_Dtdij> _ £>, , , . . _ c
Denoting the surface area as A,, the boundary leakage from one surface of a volume element is given
by
Ls,K=-D,An (11)
\
JAKKI 12(>0 :•,. IVt.-uU ..I tin- M.-ili.
80UNDARIES
1
1 —-o—
i
i >!^j 0tj I
1 1
i i
1 ^ \ i
i° • 1 . • L-|-k
by d e f i n i n g (he f l u x v e c t o r as
M =
A Cn-iirial Dinifnr-i.inal Ni-imnn UifluMuii < "--il«.-i il.it i- .ti C . l r : AIX' 1 \I-.RI 12HI)
0, . ,n 1
0, .
o. II >
A
2 " • • £ • ] •
/•' =
V'TI-
In tlu' finite difference approximation sli'uvti in I lie [irm'tlin/f seel ion, Ivq. (\'±) is romcrlerl ID ;I
set of algebraic equations:
where .1 is the algebraic matrix operator, which contains the diffusion, scattering and loss operators.
As the largest eigenvalue of Eq. (!.'!) is physically significant, the solution to Eq. (l.'i) is expressed as
For obtaining the solution of Fq. (13), it is necessary to invert the matrix A. For most problems of
interest, however .1 is too large to invert. In a case which involve for instance, 10,000 space-eneigy
points, the matrix A contains 100,000,000 elements, which are far too many numbers to handle efficiently
with present generation computers. Thus in the ADC program, two matrix iterative techniques, either
of which may be selected by user's option, are prepared for solving Eq. (l.'i).
The values of neutron flux as for all energy groups are initially guessed and the fission neutron
source for the first iteration is calculated using this flux guess. The fission neutron source for the
m-th iteration is defined by
where 5i»",j,i is the (m —l)-st iteration flux of the g-th group at mesh point (I,J,K). With this fission
neutron source, X, S(,™,i, one can obtain a set of multi-group fluxes, beginning at the first group and
carrying on down to the thermal (N-th) group. As the slowing-down source in Eq. (7) depends on the
fluxes, it is updated every time of iter?.cions by using the newest flux values. This iterative process
is called as the inner iteration and is performed by the subroutine INNERl or INNER2.
The finite difference equation (<?) is rewritten in the matrix form:
A,4,=S, , (16)
JAKKI 12i>0
( V -'. 1. I •%. 2 , 1 . 1
V'."> I i .ViS 1
( l
i'J- V j . .VI/, \t \ . ,V.J. .Vt. ,M
The point wise neulmn source •!>„,,,,,,, wliicli consists of fission ami slowing ilnwn sonriv -M Ihcnicsli
IKiint (.i.J.K) in tin- p-th Kn>up, is lhe m-lli I'lonifiil of tIJI- vcrtor Se with
whero A',, A', and A'^ aro the maximum numbers of mesh points along X. Y and Z axes, respectively,
in a similar fashion, the matrix .!„ is constructed with elements A,,} defined by
11';
The iniK'r iteration process is accelerated by successive over relaxation v^Oiv) or successive line over
relaxation (SLOR) method. The iterative scheme of SOR is represened as
•=S,.... , - C u . , , j . ,{\'^ ^ 1. + C4.,.J.1d|'"i'i' t 4 C 5 . t . ,, ,<;•,'•••• ( , ^ C,-.. ,. .. < ^ , " ' J ' , ' ) . L'l
The .solution of Eq. (21) involves the well known problem o[ inverting a Iri tlianon;\l matrix by (i:iuss
reduction technique.
The convergence rate of an iterative process depends on the ivi.:\al ion laclor. In the case of SOK
of SLOR method, the spectral radius, p. of the point or block Jacobian matrix \e.ti. for SOR. I he
|X)int Jacobian matrix is L + V) is needed to determine the o p t i m u m relaxation factor;
• ) • >
-p"
For estimating p by using the in th iteration flux, ci"' .which is in deiHiidenl on the source
vector S.j, the source vector is set to zero and the ixiwer iteration for findiiv,' the largest eigenvalue of
Jacobian matrix is performed in advance of the inner iteration process. However, since fairly lime
consuming iterations are required for estimating/), the total running time is not always reduced with
the use of the calculated optimum relaxation factor. Thus, in ADC program, choice is permitted whether
tIv.1 constant relaxation factor is given by user's exjxTienec or the optimum factor is computed from
the simplified spectral radius, />, of the Jacohian matrix. In the case of SOR, this s|ieclrai radius is
given" by
where N is the number of dimensions and /„ is the number of mesh intervals in the n th direction. In
the case of SLOR, we omit, from the summation in Eq. (,'2;o, the dimension along which the line
relaxation is done to give
The inner iteration is repeated unlii the following convergence criterion is satisfied.
where S " ' " " is the ( m - l ) - s t iteration fission source. The eigenvalue, and its upper and lower bounds
at the m - t h iteration are defined respectively as follows 51 :
, i: s .s'"'
im
'" ~ '•'•' v; s .s"""
i . J , Jt t. J . k t. J . k
s'm , ,,
K = M a x [ ;•„/_•,';}• H S'"'\ . (29!
•'•' ,-....,• S '•'•" '•''
i. J. k
JAI-1KI l a i O :!. I V i . n l > •>( l l u M . - l l i . « l 9
S'"" in ,,
in which n is j»iven by
Ui <!.v)i'<w"., ">
where /i,,. <i, and Ji.v are the largest, the second and (lie least eiui'inalnes of the matrix /'. The
envaluo Xx is assumed to be zero and the dominant ratio. /, ,'^0. is approximated by
, A'
*' = l - ^ « . CM!
(7
" j . i i. j . * i."7t * i. J. t
1 + V 1 -JVfax^-,',] 1 +V 1
One-through-iteration
The iterative technique described below is similar to that used in the EQI'ITOISE program. In
this method the convergence of neutron fluxes at all space-energy points is tested at the end of one-
through iterative step. This iterative process may lie described as
?5<->=:£f<»'5l<»-'> , (37)
where 0im> is the flux vector for m-th iteration and defined by
W A l < « l Diim-iisinml Nwiin.n DIIIIIMIUI f.iUu1.itI.HI l'.«l. AIK' 1AKK
,}<-> = , •
where <!>./"" is the subvector shown in Kq. (16), and B"" is the iteration matrix which is varied with
iteration because the fission source is re normalized at the end of a sweep through all mesh points
and all groups. The matrix form of Eq. (37) is applicable in lxith cases of point and line relaxation
method. Kor example, if point relaxation is used according to Eq. (18), the subvvclor of any j - t h group
for the m th iteration $,""> is given by
^ V " — (l~ <*>/-.?) M(l — <w)/+ «^M;}^V" ' J +w(I — (oLg) '
y i in > *V A
' i » IT; I I
Equation (38) can also be written in the matrix form ;is shown1 in Ki|. (,37). The most straight
foward method for solving Eq. (38) is basically the [lower method, which is applicable to any matrix
whose largest eigenvalue is unique. Kor B'"" with such a form as the discrete approximation matrix to
neutron diffusion operator, its spectral radius is bounded101 by
cmj (ffl)
/l <
Min -"• 1; i<p(B "")< Max -°«L'/,— • (•'(>)
». t. ]. t «. i. h k
where ^ m J j ,. is the m - t h iteration flux of the g-lh group at the mesh ixiint ( l . J . K ) and p(B"") is
the siK'ctral radius of B'"'\ The |0(B 1 "") must tend to u n i t y as the iteration proceeds because of the
fission source r e n o n n a l i z a t i o a When the flux convergence criterion ( w i t h an input parameter £,)
m)
$ 5*
Max Min •'•-,';•"•-I (•11)
'. >. i. ' ' , J. *
J> t
l r '& I I^ l
^ ' m ) = - i - Max j-ftji^i*' + Min ;;'•/;'" . (.14)
\ f, 1. /. I ' v ' Ji I. i. I
This extrapolation is not applied to the first /„ iterations and after ^,-th iteration, the change
JAEK1 12()O :i. Di-t.iik nl ilio MI-IIKKI n
of the extrapolation factor is tested every Ln iterations. The /„, and /.„ are sin'cified as input data
and the extrapolation criterion is given by
i _ v ^*o V,, V
^'J-0 I • *-tf I . P ^ T - *-/.O 1 .
A,//
/.,.„•--?« i/i-/..' <„' •••„ or g - 1 ) , tff.ff'=1.2, N) .
The adjoint operator /.* to Kq. (4fi) and its solution, i.e., the adjoint flux $!« must satisfy the
following relation:
1(5* . / ^ , ] - i ( i . /,*,5»1 . (17)
Therefore, we got
- F « » r ^ / + ^ . « ^ * - i ' . . . n i S A , i + - ^ - ' ^ , ( i ; . (g-1.2, /V) , (18)
or in the matrix representation.
4- Program Usage
The required core memories .ire depend sensitively on the problem to lie solved, since the ADC
program solves either one- two- or three dimensional problem, optionally. Ordinarily, by the use of
variable dimensioning, the program permits the optimum use <if available storage. However, the
compilers do not allow variable dimensions in the main program, ami hence a fixed common length
must be occupied prior to the program loading.
Since the program flow path depends on the problem to IK-solved, for an effective use of the core
storage, the subroutines which are unnecessary for solving the problem should IK' unloaded ami the
"CALL" statement calling such subroutines in the main program should also be deleted. I'ormelly. in these
cases, the overlay structure was changed, but for mull i process computers, the occupation of nnneces
sary core memory or loading of unnecessary subroutines is not economical.
The ADC program has eliminated these demerits by using two step Fortran job. A common area
and program flow path required for solving n particular problem are generated every time before the
execution starts, Tliis improved flowchart is shown iii Vig. 3.
Library
WRITF STANDARD
READ STANDARD FLOW PATH
FLOW 8 TREE
STRUCTURE on
8 TREE STRUCTURE from TAPE I
TAPE 1
I PRE-ADC END |
1
I READ Relocotoble ^COMPILE
I WRITE R.B. DISK 3
1 Binary from DISK 3
fa LIED
[EXECUTION START I
DATA EXECUTION
In prc-ADC. the input data are read in with I N l ' l ' T CARD () to INl'l'T CARD 7 (see Section
•1. 3. A) for the determination of COMMON length and flow path control. The standard main program
of ADC is read from a library tape and updated according to the input data from pre ADC. and hence
tile free structure is generated for the linkage edit corresponding to the new flow path. The BCD image
of the new main program and tree structure are saved in DISK 1 and DISK 'I. rcS|X'clively. Next,
the new main program and other subroutines are compiled and the linkage edit is made according !o new
tree structure. After reading Die remaining input data (see Section I..1. B ' . the execution starts. The
ADC program can he used also with a fixed common length for a lilK words machine as usual.
The standard ADC rode, the main program of which has a fixed common length, is divided into
five segments as shown in Fig. 4. Table I lists the functions of ADC subroutines. Kach stage of the
program fits in a computer having at least (i IK words core slorage, and uses 5 scratch tapes depending
on the problem to be solved. Table 2 gives the logical tape numbers 1 hat are referred lo in the program
and the fund ions of the (apes.
MAIM
WAP
OUT
COW Level 1
The execution time required to solve a problem is very dependent on the problem itself. The use
of a large capacity storage without using scratch units is best to achieve the minimum execution time.
However, as the core capacity is limited, an efficient use of auxiliary hardware with sophisticated
software is necessary to reduce the required core memory. In a sample problem shown in the appendix
with X-Y-Z geometry of 3840 mesh points and three energy groups, the amount of the required core
memory for the ADC code is 108K words, on the other hand tor the CITATION code, is 207K words.
This result shows that the way of the selection of common area and program flow path of the ADC
code is effective to reduce the required core memory. In this case, the execution tim>. for the ADC is
79 sec and that for the CITATION is 77sec.
Except for the control parameters, the cross sections, spatial flux guess and all other floating point
numbers are read into the ADC program in the free format via generalized subroutine REAG. The
JAI'KI I2«) I. IWram t V w J.,
subroutine RKAG for KACOM 230-75 converts BCD informal ion to integer or floating point binary
information. Three conversion types of reading N floating numbers, N integer numbers, and !>N,
characters, N2 integers and N 3 floating point numbers are allowed by convsp>nd:rig subroutines. r"
A typical example of Ihe function of this subroutine is described for the following three punched
cards;
105. :118. - 1 1 , 1.5E-3 :i. 1 2 E - 3 / THIS IS A COMMENT
1.0, 5»t>. 1, 2 . - 0 . 2 /
The subroutine will convert UCD number 105 and !US to their binary integer equivalents. In a similar
fashion —M will be eom/erled to a negative binary integer, and 1.5K — 3 will be converted to (1.(1(115.
Data punched on a card may be delimited by blank column or comma The slash (, ) indicates the end
of the BCD field to be converted, and the remainder of the card can be used for comments. If no slash
is present, 72 columns of a card are scanned and the next card is read.
The second card indicates that Hie data 1.0 and 1.5 in the first pamuheses and I) in the second
are re|>eatod twice and three times, respectively, and therefore it is equivalent with a card punched
as 1.0, 1.5. 1.0, 1.5, 0,0,0/. The last card inrlicales that the first word is 1.0. the second word 0.1
is successively accumulated five times to the previous word and the last word —0.2. twice. Therefore
it is equivalent with a card punched as 1. II, 1.1, 1.2, 1.3, l.-l !..">, l.:i. 1.1/.
The format for the control parameters (CAK1) 0 — L'AKM) 7) must be 12I(i for integer, •>Klli. 5 for
floating point data and ISA I for job title and comments. The input data from CAM) 0 to CARD 7
are used for determination of COMMON length and flow path control.
Word Name of
Number on Card Variable Comments
Number Name of
Word on Card Variable Comment
The following data are read by the subroutine REAG. The formats denoted by I and E mean
integer and floating point number, respectively.
NKR(NFR*2) I NFR*2 Required only if NFR > 0. Energy group number for flux ratio
calculation (e.g. if NGMAX = -1, NFR = 2).
1, 4, 2, 4/ In this example, <f>i/$,, Qz/'t't pointwise two ratios
are calculated)
IGl(NBETA) I NBETA Required only U NBETA > 0. Energj group number to be
normalized (e.g. if NBETA = 2, NGMAX=4).
3,4/ In tlrs example, 3rd group, -1th group fluxes are normalized
by input specifications. (Other group fluxes are normalized by
total power.)
BETAF(NBETA, E Required only if NBETA > 0. NBETA blocks data are required.
NRMAX) Each block has NRMAX entries.
1st block; IGl(l)th group flux normalization factor.
2nd block; IGl(2)th group flux normalization factor.
JAERI 1260 4. Program Usit
4. A Output information
Output data of "ADC" consist of two parts, normal edit quantities and optional regionwise edit
quantities. Varieties of integrated and averaged values for specified mesh regions, which are not always
identical with material regions, are edited by the edit segment ADCED.
The follwing is the list of normal output quantities. All pointwise edit data are normalized so
that total fission source equals to unity.
(1) All the input is printed immediately after read-in.
(2) The two dimensional material map by integer numbers, (for three dimensional problem, multiple
maps corresponding to all X-Y or R-0 planes)
(3) Eigenvalue convergence status at each iteration.
(4) Gross neutron balance.
(5) Pointwise group flux.
(6) Pointwise power density.
The following quantities are edited optionally if the user requires the integrated and averaged
values in edit-regions or the re-editing of the result saved in tape.
(1) All the input of "ADCED" is printed immediately after the reading.
(2) The two dimensional edit-region by integer numbers (for three dimensional problem, multiple
maps corresponding to all X-Y or R-0 planes).
(3) Pointwise edit
(a) Pointwise flux 0S, (, s, t
Renormalized by the formula
if N B E T A = 0 , / 9 - i = s [ is constant.
If N B E T A > 0, Po (NR) is specified by input for the edit-region for each energy group
where NR : edit-region number
PCONV : conversion factor t o power
PT : total power
(b) Pointwise flux ratio aXFS,,, j ,k
Ojf FR. (. i, t ~$'t, I. J. */0 V . (, 1, *
where neutron group number g is specified by input.
(c) Pointwise power P,, x, t
22 A General Dimensional Neutron Diffusion Calculation Code ADC JAKRI 1260
These quantities are renormalized so that total neutron source; V] 2 Fvn. „ equals to unity.
.v/t a '
(e) Region averaged flux <j>NR,,
All of the flux integrals and averages are multiplied by the normalization factor.
(f) Region averaged power, PN11
9 0
(h) Edit-region volume
V
XR~ \ J , V S 2 E I ' "(. J. t
Acknowledgements
The authors wish to express their thanks to V. Kuge of the Japan Atomic Power Co. for his
valuable suggestions about the numerical method. Thanks are also due to S. Matsuura and T. Asaoka
of JAERI for their valuable discussions and their critical reading of this manuscript.
References
1) Fowler T. B. and Vondy D. R. : Nuclear Reactor Core Analysis Code : CITATION, ORNL-TM-2496
Rev. 1 (1969)
2) Hardie R. W. and Little W. W. Jr: 3DB; A Three-Dimensional Diffusion Theory Burnup Code,
BNWL-1264 (1970)
3) Tobias M. L. and Fowler T. B. : The EQUIPOISE Method—A Simple Procedure for Group-Diffusion
JAKKI 1260 4. I'muram I'KIKI- -•'
Calculations in Two and Three Dimensions, Nucl. Sci. Eng. , 12O> 5U SIS U!Mi2>
I) Varga R. S. : Matrix Iterative Analysis, l'rentice Hall Inc. (\\)(,2)
5) Akanuma M. , Kuge Y. and Yasukawa S. : The KAK Program for the Numerical Solution oi 1-i-w
Group Neutron Diffusion Equations in Two Dimensions, JAKRI 1127 (Idti7)
6) Asai K. and Tsuchihashi K. : A Subroutine Reading Data in Kree Format, JAER1 M .1-158 0971)
A < K-tM-nil D i l l l t ' l b i i i i u l Nviilnni I)i(Hi>iun Cili'uUluin I'mli-:A I X " 1AKKI laid
The following pages show the input data and computer output for li group, 2-malerial, eigenvalue
problem in X-Y-Z geometry (see Figure A- 1 ). The problem requires nlxml 79 seconds to run.
_^ MATERAL3
/vy ->
(0,66.7,01
'(533, 33 3,01
1
/
/ MATERAL I
(0,0,0) (66.7,0,0) (1076,0,0 J
'.^
l i i
,. rerr ;f£Pf-::e'^LaFLT=,«,EEiR:,- : r,V I
-^'
• - - -- ^ S - i t - t L !:-•(•: ••:,';•<Jt-, t l 4 4 e "Ef-'P--1
!
'fl4f -HRUI-i i - D C - ^ l - U T - ^ t i t ^ - ' v - T i r j ^ l s
F .Sf'LCTK
O.lOOOOt+01 0.0 O.I)
bOUNOARY
1
I MAT iO
1
2
JAEKI 12I>0 Ap|>cn/li\ ; Simple input find
H N (V r!
28 A Gonrral Dimpnsinn.il Ni-utimi Oiffii-i.nl (VikulMimi C.»li-: AIK" I A I K I l^ii
c o o
3 3 -Z-
,"\ f\ r-l
1 I I
C * .-:
r. 7 - \ .
O O
O 3
o 3 C 4 +
UJ ^..
r _. _,
c o
,-t M o o
O O C-
1 1 1 ( » CO
jJ _LJ *l
o j
er o sC O c o
•T
J •H -A
•r> o
!-• ~M —O
oo oc 3 -. Q 3
O 3
CO O 3 .-H 3
3 3
o o O 3 O
o
o o oo c; o CC CO
CM CM
. j j j -d
•H
IT\ °. °.
3 C
-
o CT
r-t
" ^ ao -*
CO CO , * • *
.Li UJ
3 — -. D —,
O 3
_ _ 3 3 :\l
O O O O «-i
3 3
i-!H
3C
J .2 Z. - •
C C O 3 -C
OO O3 O O
CO O -^ O O
3 3 * S, 3 -tf
O .U U 3 aJ iJ 3 j j .JJ 3 3 3 O
J O C ^ O O 5OC 2: (**• ?<". i m m
I O O X O O X 3 O ! X. X 0" X • » "^
n m n / p ^ m ••/•. o t*-
^j f o jj c c J C t
o *- C O
a: <
* * •
> c rH rg tn
•- •< o c. > o o
4 < i 4 4
X X X 1 X X 1 X
5 i x 1 1 1 1 1 J
•1
1 1
X 1 X X X 1 X 1 1 1 1 1 1 1
6 <
•1 1 1 1
1
1 1 X X X X 1 X X 1 X 1 1 1 J
7 <
f 1 1 1 1 1 1 • 1 1 1 1 1 I X 1 1
8 <
> 1 1 1 1 1 1 i i 1 1 1 i 1 .1. 1 1
9 *
t ^ 1 1 1
X 1 X X 1 1 1 X 1 1 1 ]
10 < J 1
> 1 1 1 1 1 1 t 1 1 1 1 1 1 1 1 1
11 *
• 1 1 1 1 1 1 1 1 1 1 1 i 1 1 1
12 H
]_ ^ ^ » 1
1 •1 1 1
1
J, J- X 1 X 1 X 1
13 <
» 1 1 1 1 1 1 I 1 1_ 1 1 1 1 1 i
14 *
» 1 1 1 J 1 1 1 1 1 1 1 1 1 1 X 1
15 <
> 1 1 X 1 1 1 1 ) I 1 1 1 A, i. 1 1
16
1TH1S HAP TO I MESH POI'.T MU
6 '' 8 9 10 11 12 13 I t ID
OX DIRECTION i',ESH * | ( ) T H
0.O83OOE+01 0.68300E+01 0.683U0E*01 O.6S3OOE+O1 0.6d30UU 0.
Otb8300E*Ul O.b83OOE + Ol 0 . ti83OOF-»Ol J.683'JOE*'J1 0.68300L O.b8300E*()l
OY' U l H t t T i U N MESH MOTH
0.68300fc»01 0.6fl300E + 01 0.683<)Ur + 'Ji 0 . r,a3U0E + 01. O.68300L*ol 0 . ' ! . f > i 3 ' J t • J l '.). t = 8 3 ' , 0 E * u l n.
0.68300E + 01 0.fc8300E + 01 0.o830r)E + j i O.683UUE + O1 0,6830ut + ul 0.
0 X 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
l l l l l l l l l l l l l l l l
l l l l l l l l l l l l l l l l
l l l l l l 1 1 1 1 1 1 1 1 1 1
1 1 1 1
l l l l 1 1 1 1
1 1 1 1 1
1 i i l l :
7
2* 1 1 1 1 1 1*
8
2 2 2 2 2 2 2 2* 1 1 1 1 * 1 1 1 1
9 * *
2 2 2 2 2 2 2 2 * l l l l l l l l *
10 * *
2 2 2 2 2 2 ; 2 * 1 1 1 1 1 1 1 1 *
11
2 2 2 2 2.
12
2 2 2 2 2 2 2 2* 1 1 1 1*
13
2 2 2 2 2 2 2 2* 1 1 1 1 *
14
2 2 2 2 2 2 2 2 2 2* 1 1 1 1 1 *
15 *
2 2 2 2 2 2 2 2 2 2* 1 1 1 1 1 *
16
Lt AK.Abfc
3.074E+01 1, .V?- L-C'J •J t L " W 7 I * )>^f — t. 9 •!. jilt-'i'-' J • ! Vfc£, ~f J ^ I. ,j i* J* f — u 9 0 d.5blt-10 4.192L-10
fi.^4>.E-09 7 . D ii6e-u9 5.B v4F. - U 9 4 , t 2 1 E - ' l 9 3 . i 9 t i t - 0 > ? . /int-u9 1.949E-09 1.31ot-09
3.75 7(.+(.l fc 1 .61 1L-0'.' i.*9^fc-fi* «f-UB 1 -0' i ) : —L D fl,f,3'J^-09 f . r , / ' 9 [ . - 0 a 5 . i 94t-09 J.7JBL-O9 /.SOit-O"
4,440E+01 7 •/.,b5(- L-CH.- ?.'*bOt-('» •)0f -od i • H''** f- — u f i i . '' 2 f 't — t ^ 0 1
••
. * 1 *H — t; f* 9 .3** "*F —0 9 6,Mufc-U9 4,47rk-09 ?.M4f-0-<
5.1231.+01 fc 3 .<»f". h-jf, 3. /cbt-O- .•>. J.,3!--OB i'.9 ?.4i;?fc-'in t . o l 2 t - t . ' 1 . *>•>' t - l - t "
-l.fl! • " •
i.l<?Jt-!;ii (".ftlOt-09 4,,,;7t-C9
5.80bF-+01 9 5 . b o k t-0>. 5.i63E-Of '•,^ jze-oB 4 . 2 . 'f- - y d j.^eot-c 1 3 . v "l ^ t -'.- ' ? .j ^ .. ^ —U8 i,710fc-0» l,12jt-0K ft,7J>,t-0»
6.4B9F +01 10 7 .US'- L-(,', 6 . b 3 j L - .> f~ t'.3 • .3K-0C 5.7 -01 .lfq(.7t..,;h - . i t1 *?; — 1' ** ^ .^c! t -ii-i / . 1,ibt-:)<* 1.35tit-'JB 9.3*9L-09
7 . ?. 7 2 F-. + 0 i J- J (i . 6 1 3 L-fjl- (J.HSUE-OH 7 .;,
44K.UP 7 . 1 " • ( - - t . ' 1 ti,2 3^F-i'n i . ^ 3 ; , t - ' . " < * • . I'l/.'-Of
7.H55L + U 1 't- i^ • 0 0 !
L-07 *.'i'23F-ns 9. i r ' ' L - O B f • 3 / b t- I . , b , ' . 4 l ( - t - ( l f t . / ? - [ — • ' ' ' . l,320fc-0B
1.3
8.»37h+t)J j ..II-.'t.-0' 1.06Jh-07 1 . 'JJlf.-Of' 9 . u f ft- - i , n ,1, ?<fct-i 1 'E-Od
Appiiulix :Simple mpiii . u n l <uiijiui
JAKKI la
33 ac co
o o o
I 1 I I I I I I I I I
•-' —* s :"5 3 T ?• J* 7* 3* .c * o -o
o o c
r r <i i i i r i i i t I I
—IO3O£T>'>a'>3s"0'0
—i .-I —< —I O O O -J3 O : - "5
f I I I I 1 I I I I I
L^J ^r ^ i t I i * t ^Li • L ^ •> I JL -• i I I . i i I
JC f \ J \T>
a; co a>
0 o o
1 I I I I I I I I I 1 I f1
ceo
I I I
UJ id LU
UJ LU tU
1
—< •-" O O O D r- .-> "
i T7 T i i TT »
- g t v M - v - H -t ~< a a o o o o O D ^
0 c- r: O O O O O O C
1 I I I I I 1I 1I I I I 1t I I I I
I I I I I I I
J5 3 .
I I I I I I I I I I I I I I t I f O
o 1 1 I
• UJ U.
^ if
-3 *V O CT-
c3 en
-a- •>
r^ >- • - O
a 3o ^
I I I 3
• ^ iA
1 r* •-* .-* <\ O #
— JO
O O ^ O^ ^ ^ CO •£} OQ CD CD "X; X. *
*-i »H .-: «H «-i .-i »-i c* o ca c .2 o 3 o ~ */*. -« H •-» o c o — c o c c o 0 o o
I I I I I I I I I I I I I I I I O3 I I I I 1 I I I I I I 1 II
•o % JC
»-l»-4OOC'O0 0 C OOC _?C. OOOO • c e o
I I I I t I I
I r i i i i i i * i i i uo •i. U -U
i <r o -c
! 1-| -JD »H T-i
0003000030303 ^ ; ^ 3 30033033303
9.903E+01 15 f.9b4L-0t' 6.790E-0B 6.447E-08 5.947E-0B 5.311E-0B 3.73?E-O8 2.874E-08 2.05bE-0B 1.354E-08
1.O59E+O2 16 7.l41t-DH fr.964E-0S 6.616E-0B b.U0E~OB 5.464E-08 4,705h-08 3.bb9E-08 3.OO5E-O8 2.W5E-08 1.450t-08
2 GHOllP FLUX. I POSITION O,40000t+Ol
X 7.172E+01 7,e55t+Ol 8.537E+O1 9.22ot+Ol 9.903E+01 1.039E-02
11 12 13 14 15 16
3.415E+CU 1 1.20f!£-ll 7.191E-12 4.075fc-12 i.l9BF-12 1.1UE-12 4.t>03i-13
1.O25E+C1 2 3.11bfc-ll 1.B30E-11 1.023E-11 5-438E-12 2.713E-12 1.111L-1?
1.708£*01 3 b.7l7t-Ji 3.87SE-11 2.126E-11 1.109E-11 5.438E-12 2.19Bt-12
2.391E+01 4 1.369H-10 7.747E-11 4.1bOE-ll 2 . 1 2 b F - l l 1.023E-11 4.075E-12
3.074E+01 5 2.fcb0t-10 1,'»7<,E-1O 7,747E-11 3.O77E-11 1.830E-11 7.19Ot-12
3.757E+O1 (, 4,89BC-J« ?.66oE-10 1.369E-10 6-717F.-11 3.116E-11 l , 2 0 8 t - l l
4.440E+01 7 B.474E-10 4.515E-10 2.281E-10 1.101E-10 5.032E-11 l.-«lt-ll
5.123E+O1 8 1.365E-09 7.l6hE-10 3.572fc-10 1.702E-10 7.704E-U 2.934fc-ll
5.8O6E+O1 9 2.03<!t-09 1.061E-09 5.255E-10 2.490E-10 1.122E-1O 4.?57f.-H
b,489E+01 10 2.«31b-O4 1.477L-09 7.319E-10 3-V7i>t-10 1.564E-1" 5.934t:-ll
7.172E*C)1 11 3.723^-04 i ,95^E-n9 9.7olF-lO '(.665E-1U ?.109E-10 ».015t-ll
7.B55E+U1 12 4. 721i:-O9 2.518t-09 1.271E-09 b.lUlE-10 2.768t-10 1.O53L-1O
8.537E*0j. 13 5.644E-09 3.1b4h-09 I.tii2t-U9 7.7fbE-10 3.531E-1O 1.3'»3L-10
9.220E+U1 14 7.04hi_-09 3.B6/)E-oo 1.9^9E-09 9.56UE-10 4.334E-1U 1.644E-10
9.9O3E+01 15 8.141F-09 4.497E-09 2.3U9E-09 1.H3E-U9 5.031E-JO 1.902L-10
1.059E*02 If, «.791t-09 4.875E-09 2.5O3E-09 l-2o?E-O9 5.442E.-10 2.L/53fc-10
3 OKOU^ FLUA . L POSIT OM 0 .40000E*01
X 3.41st+yo l.(;25L>01 l .70UE-HJ1 2.391E*01 3 . 0 7 t t * ' ) l 3.757t*01 4, 4«0E+Ul 5. 123t*'Jl 5. (JObOOl b.4o'Jp + o l
1 2 j 4 5 6 7 •• 10
Y
3.-U5E + 00 1 3. 44Hf -o1) 3.199E-09 2 . ' f 2 E - 0 9 2.272E-O9 1.788E-09 1 . 364t.-09 1 .012E-O9 7 .27it-iO 5 • 017E-10 3.: J 03fc-10
1.O25E+O1 2 1.0fl3t-n-i 1.004E-08 8,bh2K-09 7.11OE-O9 5.59SE-09 4.279(.-09 3 •lolE-09 ,2bbE-J9 1 ,573E-09 1.C.30E-09
1.7O8E+01 i 2.4l«£-oi /.*3'>F.-0f> '. .'O'tt-08 1.5«4E-08 1.250E-00 9. j 9 7 t - 0 9 7 • lc^E-119 .1^1E-O9 2.291E-09
•1 i r t \ ^
5 3 •531E-0V
£i391E+O1
I I I
4 4.94c:t-0« - . ' 7 3 E - O 8 J.-J54E-08 J.Z^oE-OO 2.579E-O8 1.994L-01< 1 ,4^7F-06 1 .0 'SE-08 7 ,35Bt~09 't, 724E-U9
r
3.O74E+01 5 ). ton i-a* k • '2«E-0-f 7.5 /8E-0D b.2f3E-on 5.032E-OH 3. IHr -i'S 2 .975E-O8 ^ .14 fE-On j , « 5 b t - 0 S ••(.^lE-CV
3.757E+01 h i . b 5 V L - f l ' 1.54bE-07 1.35SE-U' l.lj/E-lW 9.25bE-oa >".33<?L-0H 5 .5S>5F-oe .Onbt-uW . 'J3t-0s 1 . 71-t-l^S
4,440t+0i 1 2.bo5L.-n ' 7 . 5 2 3 E - O 7 ?.?ttl-bl i . 9 l B f . - o 7 1.591E-0 7 i . ^ 7 i i L - 0 r 9 .153E-OB 7 .141E-08 4 . ?9SL-0-- ^ . ^ 7 7t-0B
5.123L+O1 » j . 9 a n.-o' 5.7S4E-07 J."j8t-u7 2,9^3^-07 2.5JZE-O' 2 . J 6 b r - U 7 1 ^04t-07 x ,lt>bE-07 7.60H-C '», 79sE-Ofc
5.8O0E +01 9 5 .476t-0' s. <:4<)E-0/ 4. o.lbE-0 1 "•3J3E-J7 3.701E-O' 3 . o 5 7 t - O " 2 .3-»2F-07 1 .744E-07 1 . lbbt-07 7.141E-C8
10 7
b."B9E+0i . 0 3 4 t - 0 1 6.7R9E-0'' O.332F.-O7 5. 7IlE-'>7 •..971E-O7 4.14«K-0 7 '1,;fr't-07 ^ . 392E-U7 1 ,f,05L-3 7 9,o53t-0o
7.172E*O1 11 8.54'3^-0' H.2b4E-l)7 7. 7&5E-O7 7.0B3E-O7 h.2l7E-07 5 . 2 2 2 t - 0 7 i* .l-«>t-G7 3 .057E-07 2 .OobE-U 1 1 ,28Of-01
7.855E+O1 lc 'J.919t-0"' V.b«tlf.-()7 y . i O l E - O f l).3 1 '8E-j7 7.353E-O' 6.217f.-07 4 .'/71E-O7 J ,7iilE-07 2 ,53<ic-nf 1.5UE-C7
B.53 /E + 01 13 i . 1091--Ob J .OoOE-Of. j.J22E-O6 V.3h7E-07 o.32^E -07 1 .u83t~C7 5.-lit-07 4 .3u"t-07 .99jt-07 1.918L-O7
9.22UE+U1 14 1.201t-0o 1.170E-0b l.HOb-06 1-O22E-Ub T.lOlE-0'' '.7B5L-07 .^J2E-07 "* .3Jbt-07 . 4 2 o t - 0 7 2.244E-C7
3
9.90JE+01 15 1.2b4h-0t 1.232L-Uh l.WOE-Ob LOBOE-Ob 9.641E-0' B. 26»t-O7 t .7o9E-07 5 .249E-07 3 .785L-C7 2 . 5 2 3E-C. !
1.059E+02 16 1.29bf-0b l.iti'f-S'i 1.201E-0b 1.11/9E-U6 •S.919E-07 d.54^L-07 7 .^34E-07 5 .4 fbE-07 3 ,987t-07 2,b85t-C(r
3 6(<0u>' f-LU* • L P U ' J I T 1ON 0 .4OU0Lfct01
X 7.172t»01 7.J55!"-0.t « ,537fctOl 9.220£*01 9.903E+01 1.059£-U2
11 12 13 1* • t
3.72bE-ll 1.832E-11
FLUX* /. voanwi
9 . 9 0 3 E +0 1 l.uiVt
11 i j 13 16
3.75-fK-J)
1.706EKU :.ii3fc-:" 3.72f,t-n
].622E-U9 i,7'»2t- • l U
;
3,<mf:.<-0i 3.06U-09 L.O22K- •US) 3.758t-l ' l.il'H-W)
3.757E+U1 6 l.OOJr-On b^iiTE-lU 2.U66L-10
7 1.030fc-3» l.lUSt-lO
9
5.806E+01 9
6.489E+01 10 3.131E-C * :.jl2t-u^
7.172E+01 11 7 , 333t-Uti 0 fl
3 . *t 3 31 " 1 , ^ ^ ^ f*. — u 8
Y.855E+O1 12 S ,U3 ^E- .579E-O8
8.537E+01 13 1 . 137, -07 6 .27'»L- 0 8 3 .248E-08 9 2.
9.220E+01 14 9 2.
9.903E+01 15 t.0<Mt-Q3 3.
1.059E+02 16 J..659t-Q7 4.910E-08 a 3,