0% found this document useful (0 votes)
15 views33 pages

CFD강의자료5 Ch4

Sogang University CFD lecture 5

Uploaded by

이동근
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
15 views33 pages

CFD강의자료5 Ch4

Sogang University CFD lecture 5

Uploaded by

이동근
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 33

The Finite Volume Method for Diffusion Problems

An Introduction to CFD: the FVM, Ch. 4


H.K. Versteeg & W. Malalasekera

전산유체역학 강의자료 5
Dec. 2022
N. Hur
Dept. of mechanical engineering, Sogang University
4.1 Introduction
Differential and integral forms of the general transport equations

(  )
+ div( u) = div( grad  ) + S (2.39)
t
Rate of increase Net rate of flow Rate of increase Rate of increase
of  of fluid + of  out of = of  due to + of  due to
element fluid element diffusion sources

(  )
CV t dV + CV div( u)dV = CV div( grad  )dV + CV S dV (2.40)

 div adV =  n  adA


CV A
(2.41)

 
   dV  +  n  ( u ) dA =  n  ( grad  )dA +  S dV (2.42)
t  CV  A A CV
4.1 Introduction
 
   dV  +  n  ( u ) dA =  n  ( grad  )dA +  S dV (2.42)
t  CV  A A CV

Net rate of Rate of increase


Rate of + decrease of  due to = of  due to + Net rate of
increase of  convection across diffusion across creation of 
the boundaries the boundaries

 n  ( u)dA =  n  ( grad  )dA +  S dV


A A CV
(2.43)

 
t t  CV (  )dV  dt + t A n  ( u)dA dt
(2.44)
=   n  ( grad  )dA dt +   S dV dt
t A t CV
4.1 Introduction

Steady state pure diffusion problem

(  )
+ div( u) = div( grad  ) + S (2.39)
t

div(grad ) + S = 0 (4.1)

 div(grad )dV +  S dV =  n (grad )dA +  S dV = 0


CV CV A CV
(4.2)
4.2 FVM for 1-D Steady State Diffusion

d  d 
 +S = 0 (4.3)
dx  dx 

Fig 4.1
4.2 FVM for 1-D Steady State Diffusion

Step 1: Grid generation

Fig 4.2
4.2 FVM for 1-D Steady State Diffusion

Step 2: Discretization
d  d   d   d 

V
 dV +  SdV =  A  −  A  +S V = 0
dx  dx  V  dx e  dx  w
(4.4)

W +  P
w = (4.5a)
2
 + E
e = P (4.5b)
2
 d   E − P 
 A  =  e e
A  (4.6)
 dx e   x PE 

 d   P − W 
 A  =  w w
A  (4.7)
 dx w   xWP 

S V = Su + S pP (4.8) : Source Term Linearization


4.2 FVM for 1-D Steady State Diffusion

 E − P   P − W 
e Ae   −  w Aw   + ( Su + S pP ) = 0 (4.9)
  xPE    xWP 

 e w   w   e 
 Ae + Aw − S p  P =  Aw  W +  Ae  E + Su (4.10)
  xPE  xWP    xWP    xPE 

aPP = aW W + aEE + Su (4.11)

w e
where aW = , aE = ,
 xWP  xPE
aP = aW + aE − S P

Step 3: Solution of Equations


4.3 Worked examples: 1-D steady state diffusion
d  dT 
k +S = 0 (4.12)
dx  dx 

Example 4.1 : Source-free heat conduction in an insulated rod

K = 1000 W/m.K
A=10x10-3 m2

Fig 4.3
d  dT 
k =0 (4.13)
dx  dx 

Steady state temperature distribution?


4.3 Worked examples: 1-D steady state diffusion

Fig 4.3
 ke kw   kw   ke 
 A + Aw P
T =  Aw W
T +  Ae  TE (4.14)
  xPE  xWP    xWP   xPE 
e

Discretized Equation for nodal points 2,3, and 4:


aPTP = aW TW + aETE (4.15)
k k
Where aW = , aE = A , aP = aW + aE
x x
Su = Sp = 0
4.3 Worked examples: 1-D steady state diffusion

Fig 4.3

Discretized Equation for nodal point 1:


 TE − Tp   TP − TA 
kA   − kA  =0 (4.16)
 x   x 2 
 k 2k   k   2k 
 + = + +
 x  P   E 
A A T 0.T A T A  TA (4.17)
x x  x
W

aPTP = aW TW + aETE + Su (4.18)

k 2kA 2kA
where aW = 0 , aE = A , aP = aW + aE − S p , S P = − , Su = TA
x x x
4.3 Worked examples: 1-D steady state diffusion

Fig 4.3

Discretized Equation for nodal point 5:


 T −T   T −T 
kA  B P  − kA  P W  = 0 (4.19)
 x 2   x 
 k 2k   k   2k 
 + = + + (4.20)
 x  P   x  W 
A A T A T 0.T A  TB
x x
E

aPTP = aW TW + aETE + Su (4.21)

k 2kA 2kA
where aW = A, aE = 0 , aP = aW + aE − S p , S P = − , Su = TB
x x x
4.3 Worked examples: 1-D steady state diffusion
Table 4.1

300T1 = 100T2 + 200TA


200T2 = 100T1 + 100T3
200T3 = 100T2 + 200T4 (4.22)
200T4 = 100T3 + 100T5
300T5 = 100T4 + 200TB
4.3 Worked examples: 1-D steady state diffusion
 300 −100 0 0 0   T1   200TA 
 −100 200 −100 0 0  T   0 
  2  
 0 −100 200 −100 0  T3  =  0  (4.23)
    
 0 0 −100 200 −100 T
  
4 0 
 0 0 0    
−100 300  T5   200TB 

 T1  140 
T   220 
 2  
T3  =  300  (4.24)
   
T
  
4 380 
T5   460 

Fig 4.5
4.3 Worked examples: 1-D steady state diffusion
Example 4.2 : 1-D Steady Conduction with heat source
L = 2 cm, q = 1000 kW/m3

d  dT 
k +q = 0 (4.25)
dx  dx 

Fig 4.3

d  dT  Fig 4.6

V
k
dx  dx
 dV +  qdV = 0
 V
(4.26)

 dT   dT  
 kA  −  kA   + qV = 0 (4.27)
 dx e  dx  w 
4.3 Worked examples: 1-D steady state diffusion

Fig 4.3
  TE − TP   TP − TW  
 e   x  w   x   + qA x = 0
k A − k A (4.28)
    
 ke kw   ke   kw 
 A + A  P 
T = A  W 
T + A  TE + qA x (4.29)
  x  x    x    x 

Discretized Equation for nodal points 2,3, and 4:


aPTP = aW TW + aETE + Su (4.30)
kA kA
where aW = , aE = , aP = aW + aE − S p , SP = 0 , Su = qA x
x x
4.3 Worked examples: 1-D steady state diffusion

Discretized Equation for nodal point 1:

 dT   dT  
 kA  −  kA   + qV = 0 (4.31)
 dx e  dx  w 

 T −T   T −T 
ke A  E P  − k A A  P A  + qA x = 0 (4.32)
 x   x 2 

aPTP = aW TW + aETE + Su (4.33)

kA 2kA 2kA
where aW = 0, aE = , aP = aW + aE − S p , S P = − , Su = qA x + TA
x x x
4.3 Worked examples: 1-D steady state diffusion

Discretized Equation for nodal point 5:

 dT   dT  
 kA  −  kA   + qV = 0 (4.34)
 dx e  dx  w 
 T −T   T −T 
kB A  B P  − kw A  P W  + qA x = 0 (4.35)
 x 2   x 

aPTP = aW TW + aETE + Su (4.36)

kA 2kA 2kA
where aW = , aE = 0 , aP = aW + aE − S p , S P = − , Su = qA x + TB
x x x

17
4.3 Worked examples: 1-D steady state diffusion
Table 4.2

 375 −125 0 0 0   T1   29000   T1  150 


 −125 250 −125 0 0  T   4000  T   218 
  2    2  
 0 −125 250 −125 0  T3  =  4000  T3  =  254 
        
 0 0 −125 250 −125 T4   4000  T 258
 4  
 0 0 0 −125 375  T5  54000  T5   230 
4.3 Worked examples: 1-D steady state diffusion

Comparison with the analytical solution


T − T q 
T = B A + ( L − x)  x + TA (4.39)
 L 2k 
4.3 Worked examples: 1-D steady state diffusion
Example 4.3: Cooling of a circular fin by means of convective heat
transfer along its length
d  dT 
 kA  − hP (T − T ) = 0 (4.40)
dx  dx 

T − T cosh  n( L − x)
=
TB − T cosh(nL) (4.41)

Fig. 4.9
4.3 Worked examples: 1-D steady state diffusion
Solution

Fig. 4.10

d  dT  2
  − n (T − T ) = 0 where n 2
= hp / (kA) (4.42)
dx  dx 

d  dT 
 −  (T − T )dV = 0
2 (4.43)
  dV n
V
dx  dx  V

 TE − TP   TP − TW 
−    (TP − T )  x  = 0
−  2
  x n (4.44)
   x 
4.3 Worked examples: 1-D steady state diffusion
 1 1   1   1 
 + = +
 P   W   E
T T T + n 2
 xT − n 2
 xTP (4.45)
x x  x  x 

Discretized Equation for nodal points 2,3, and 4:


aPTP = aW TW + aETE + Su (4.46)

1 1
where aW = , aE = , aP = aW + aE − S p , S P = −n  x, Su = n  xT
2 2

x x

Discretized Equation for nodal point 1:


 TE − TP   TP − TB 
−   −  n (TP − T )  x  = 0
2 (4.47)
  x
   x/2 
1 2
where aW = 0, aE = a
, P = a + a − S p, S = − n 2
 x − ,
x x
W E P

2
Su = n2 xT + TB
x
4.3 Worked examples: 1-D steady state diffusion

Discretized Equation for nodal point 5:


  TP − TW  
0 −   x   −  n (TP − T )  x  = 0
2
(4.48)
  
1
where aW = , aE = 0 , aP = aW + aE − S p , S P = −n  x, Su = n  xT
2 2

x
Table 4.4

 20 −5 0 0 0   T1  1100   T1   64.22 
 −5 15 −5 0 0  T   100  T   36.91
  2    2  
 
 0 −5 15 −5 0  T3 =  100  T3  =  26.50 
        
 0 0 −5 15 −5 T
  
4 100  T
  
4 22.60 
 0 0 0 −5 10  T5   100  (4.49) T5   21.30  (4.50)
4.3 Worked examples: 1-D steady state diffusion

Fig. 4.11
4.3 Worked examples: 1-D steady state diffusion
o 4.3 Worked examples: one-dimensional steady state diffusion
- Solution
Table 4.6

       
  +   + S = 0 (4.51)
x  x  y  y 
4.4 FVM for 2-D diffusion problems

Fig. 4.12

       
V x  x 
 dx.dy + V y   y  dx.dy + V S dV = 0 (4.52)

               
 e Ae   −  w Aw    +  n An   −  s As    + SV = 0 (4.53)
  x e  x  w    y n  y  s 
4.4 FVM for 2-D diffusion problems

Flux across the west face =  w Aw



=  w Aw
(P − W )
(4.54a)
x w  xWP

Flux across the east face =  e Ae


 ( −  )
=  e Ae E P (4.54b)
x e  xPE

Flux across the south face =  s As


 ( −  )
=  s As P S (4.54c)
y s  ySP

Flux across the north face =  n An



=  n An
(  N − P )
(4.54d)
y n  yPN

 e Ae
( E −  P ) −  (P − W ) +  (N − P ) −  (P − S ) + SV = 0
w Aw n An s As (4.55)
 xPE  xWP  yPN  ySP
4.4 FVM for 2-D diffusion problems

  w Aw e Ae  s As  n An    w Aw   e Ae    s As    n An 
 + + + − S 
P P =   +
 W   +
 E   +
 S   N + Su
  xWP  xPE  ySP  yPN    xWP    xPE    ySP    yPN 
(4.56)
Discretized Equation for interior nodal points:

aPP = aW W + aEE + aSS + aNN + Su (4.57)

 w Aw  e Ae  s As  n An
where aW = a
, E = a
, S = , a = ,
 xWP  xPE  ySP N  yPN
aP = aW + aE + aS + aN − S p

Discretized Equation for boundary points should be modified.


4.5 FVM for 3-D diffusion problems

           
  +   +   + S = 0 (4.58)
x  x  y  y  z  z 

Fig. 4.13
4.4 FVM for 2-D diffusion problems
                       

 e e 
A −  w w
A   +   n n
A  −  s s
A   + 
 t t 
A −  b b
A   + S V = 0
  x e  x  w    y n  y  s    z t  z b 
(4.59)
 (E − P ) Ae (P − W ) Aw  +  (N − P ) An −  (P − S ) As 

 e −    n 
  x PE
w
 xWP    x PN
s
 x SP 
(4.60)
 ( − P ) An ( −  ) A 
+ t T − b P B b  + ( Su + S pP ) = 0
  xPT  xBP 

Discretized Equation for interior nodal points:


aPP = aW W + aEE + aSS + aNN + aBB + aT T + Su (4.61)

 w Aw  e Ae  s As  n An  b Ab
where aW = a
, E = a
, S = , a = , a = ,
 xWP  xPE  ySP N  yPN B
 z BP
 t At
aT = , aP = aW + aE + aS + aN + aB + aT − S p
 z PT
4.6 Summary of Discretized Eqn for diffusion problems

aPP =  anbnb + Su (4.62)

aP =  anb − S p (4.63)

Table 4.7
4.6 Summary of Discretized Eqn for diffusion problems

Source terms can be included through linearlised form:


S V = Su + S pP

Boundary conditions are incorporated by cutting the link to the


boundary side and introducing the boundary side flux through
additional source terms Su and Sp.
link cutting : set coefficient aB = 0 (4.64)

2k B AB
source contributions : fixed value B : Su = B ;

(4.65)
2k B AB
Sp = −


fixed value qB : Su + S p p = qB (4.66)

You might also like