APPLICATION OF A GREEN’S FUNCTION METHOD TO HEAT CONDUCTION PROBLEMS IN MULTI-LAYERED CYLINDERS
Urszula Siedlecka, Stanisław Kukla
Institute of Mathematics, Czestochowa University of Technology Częstochowa, Poland
urszula.siedlecka@im.pcz.pl, stanislaw.kukla@im.pcz.pl
Abstract. In this paper, the Green’s functions which are used in solving the heat conduc- tion problem in a composite cylinder are derived. The functions are obtained by the solution of one-dimensional eigenproblems and are presented in the form of eigenfunction series.
The temperature in the cylinder as a function of time and space coordinates are expressed by the Green’s functions. A numerical example is presented.
Keywords: heat conduction, Green’s function, composite cylinder
Introduction
The heat conduction problem in a composite circular cylinder has been consid- ered by Lu et al. in paper [1]. The problem has been solved by applying the Laplace transform and the closed form solution as the real part of a function is given. Like- wise, in papers [2, 3], the solution of the heat conduction problem in a cylinder was obtained by using the Laplace transform. The inverse transform has been numeri- cally determined. However, the numerical computations of the inverse transform often leads to numerical instabilities and for this reason the seeking of new methods is purposeful. The other approach to solving the problem is the use of the proper- ties of Green’s functions which correspond to the initial boundary problem of heat conduction in the layers of the considered composite cylinder. Applications of the Green’s functions to various problems of heat conduction are presented in the book by Beck et al. [4], in the book by Duffy [5] and in the book by Özişik [6].
The Green’s function method to the heat conduction in a composite two-layered cylinder has been applied in reference [7].
In this paper the Green’s functions applying in the problems of the heat conduc-
tion in finite, composite multi-layered cylinders are derived. The functions are
obtained by solving the heat conduction differential equation with homogenous
boundary conditions of the second and third kind. An example of application of
the Green’s functions to the problem of heat conduction in radial direction of
a two-layered cylinder is given. The temperature in the cylinder as a time function
is numerically computed.
1. Derivation of Green’s functions
The Green’s functions for the problems of heat conduction satisfy the following differential equation
( ) ( ) ( ) ( )
2
1 G 1
G r z t
t r δ ρ δ ζ δ τ δ θ ϕ
α
∇ − ∂ = − − − −
∂ (1)
where
2 2 2 22 2 2 2
1 1
r r r r θ z
∂ ∂ ∂ ∂
∇ = + + +
∂ ∂ ∂ ∂ is the laplacian in cylindrical coordinates, α is the diffusion coefficient, r , θ , z are the space coordinates in cylindrical coordinate system and t is time (Fig. 1).
Fig. 1. The sketch of the considered cylindrical region
We assume the homogeneous boundary conditions in the form:
1
G
20
a a G
r
∂ − =
∂ for r a
= (2)
1
G
20
b b G
r
∂ − =
∂ for r b = (3)
1
G
20
G
γ z γ
∂ − =
∂ for z = 0 (4)
1
G
20
G z
ν ν
∂ − =
∂ for z L = (5)
Moreover, the function zero initial condition held
( 0, , , ; , , , ) 0
G r θ z τ ρ ϕ ζ = (6)
θ b a r
L
z
We search the solution of the initial - boundary problem (1)-(6) in the form
( ) ( ) ( )
0
, , , ; , , , 1 , , ; , , cos
n
n n
G t r θ z τ ρ ϕ ζ g t r z τ ρ ζ n θ ϕ κ
∞
=
= ∑ − (7)
where κ
0= 2 π and
n
κ = π for n > . Substituting (7) into equation (1), we obtain 0 differential equation in the form
( ) ( ) ( )
2 2 2
2 2 2
1 1 1
n n n n
n
g g n g g
g r z t
r r r r z t r δ ρ δ ζ δ τ
α
∂ ∂ ∂ ∂
+ − + − = − − −
∂ ∂ ∂ ∂ (8)
The function g
n( t r z τ ρ ζ can be expressed in the form of the product of , , ; , , )
two 1D Green’s functions [4]
( , , ; , , ) ( , ; , ) ( , ; , )
n n
g t r z τ ρ ζ = R t r τ ρ ⋅ Z t z τ ζ (9) The function
R is a solution of a boundary problem:
n( ) ( )
2 2
2 2
1 1 1
nn
R n
R r t
r r r r r δ ρ δ τ t
α
∂ ∂ ∂
+ − + − − =
∂ ∂ ∂
(10)
1 n 2
0
n
a R a R
r
∂ − =
∂ for r = a (11)
1 n 2
0
n
b R b R
r
∂ − =
∂ for r b = (12)
while the function Z is a solution of the following boundary problem:
( ) ( )
2
2
1
Z Z
z t
z δ ζ δ τ t
α
∂ ∂
+ − − =
∂ ∂ (13)
1
Z
20
z Z
γ γ
∂ − =
∂ for z = 0 (14)
1
Z
20
z Z
ν ν
∂ − =
∂ for z L = (15)
We seek the function
R in the form of a series of eigenfunctions
nφ
nk( ) r of the
homogeneous problem corresponding to the boundary problem (10)-(12). The sought
functions φ
nk( ) r satisfy the equation
2 2
2
2 2
1
nk nk
n
r r r r
φ λ φ
∂ ∂
+ − =
∂ ∂
(16)
where λ is the separation constant. Moreover, the functions φ
nk( ) r satisfy the boundary conditions analogous to the given by equations (11)-(12). We assume the functions φ
nkin the form
( )
12( )
11( )
nk
r A J
n nkr A Y
n nkr
φ = λ − λ (17)
where λ
nkare roots of the eigenvalue equation
11 22 12 21
0
A A − A A = (18)
where
( ) ( )
11 1 n 2 n
A = a λ J ′ λ a − a J λ a , A
12= a
1λ Y
n′ ( λ a ) − a Y
2 n( λ a ) (19)
( ) ( )
21 1 n 2 n
A = b λ J ′ λ b − b J λ b , A
22= b
1λ Y
n′ ( λ b ) − b Y
2 n( λ b ) (20) The functions φ
nkfor each n, create a sequence of orthogonal functions, i.e. the functions satisfy the condition
( ) ( ) 0 for
for
b
nk nk R
nk a
k k
r r r dr
N k k
φ φ
′ ′ ≠
= ′ =
∫ (21)
where
( ( ) )
2b R
nk nk
a
N = ∫ r φ r dr (22)
Using the functions φ
nk, the solution of the problem (10)-(12) can be written in the form
( ) ( ) ( )
2( )
1
, ; ,
nk nk nk tn R
k nk
r
R t r e
N
λ α τ
φ φ ρ
τ ρ
∞
− −
=
= ∑ (23)
Similarly, the solution of the boundary problem (13)-(15) we find in the form
( ) ( ) ( )
2( )
1
, ; ,
k k k tZ
k k
z
Z t z e
N
β α τ
ψ ψ ζ
τ ζ
∞
− −
=
= ∑ (24)
The eigenvalues β
kare the roots of equation
( γ ν
1 2− γ ν β
2 1) cos β L + ( γ ν β
1 1 2+ γ ν
2 2) sin β L = 0 (25)
and the functions ψ
khave the form
( )
1cos
2sin
k
z
k kz
kz
ψ = γ β β + γ β (26)
and
( ( ) )
20 L Z
k k
N = ∫ ψ z dz (27)
Finally, the Green’s function G t r ( , , θ , ; , z τ ρ ϕ ζ of the problem (1)-(6) is given , , )
by equation (7) where the function g
n( t r z τ ρ ζ is expressed by equation (9) , , ; , , )
and the functions R
n( t r τ ρ , ; , ) , Z t z ( , ; , τ ζ ) are presented by (24) and (25), respectively.
The Green’s function for axisymmetric heat conduction problem can be obtained assuming n = 0 in an equation (8). That way the Green’s function for axisymmetric heat conduction problems has the form
( )
0( )
0( ) ( ) ( ) (
02 2) ( )
1 1 0
, , ; , ,
k k l l k l tR Z
k l k l
r z
G t r z e
N N
λ β α τ
φ φ ρ ψ ψ ζ
τ ρ ζ
∞ ∞
− − −
= =
= ∑∑ (28)
where
0,
0,
R
k
N
k kφ ψ , and N
kZare given by equations (18), (23), (27) and (28), respectively.
The assumption, that the heat conduction is in the radial direction only, leads to the Green’s function in the form
( )
( ) ( )
(
12 0 0 11 0 0) (
12 0(
0)
11 0(
0) )
02( )
1 0
, ; ,
k k k k k t
R
k k
G t r
A J r A Y r A J A Y
e N
λ α τ
τ ρ
λ λ λ ρ λ ρ
∞
− −
=
=
− −
= ∑ (29)
where A and
12A are given by equations (20) for
11n = 0 .
2. The Green’s functions for the problem of heat conduction in m-layered composite cylinder
To solve the problem of heat conduction in the composite m-layered cylinder by
the aid of the GF method, we use the Green’s functions satisfying boundary condi-
tions which can be obtained by suitable adoption of the coefficients, occurring in
boundary conditions (2)-(5). For example, to observe the heat conduction in the
radial direction in a two-layered hollow cylinder (m = 2), we assume in boundary
conditions (11)-(12) and in equation (30): a
1= α
1, = α
+a
2, b
1= 1 , 0
2
=
b , a = r
0,
1
1
, k
r
b = α = , to obtain the function G
1and a
1= 1 , a
2= 0 , b
1= α
2, = α
∞b
2,
1
,
a = r b = , r
2α = k
2, to obtain the function G
2. The boundary conditions, eigen- functions φ
0k, eigenvalue equations and norms of eigenfunctions N
0Rkfor that problem are presented below:
A. Boundary conditions: G
1 0 10 G
r γ
∂ − =
∂ for r = r
0, G
10 r
∂ =
∂ for r = r
1
α
= α
γ
+1 1
(a) eigenfunctions:
[ ]
1( [ ]
1) ( [ ]
1) ( [ ]
1) ( [ ]
1)
0k
J
0r
0kY
1r
1 0kY
0r
0kJ
1r
1 0kφ = λ λ − λ λ (30)
(b) eigenequation:
[ ] ( [ ] ) ( [ ] )
( ) ( [ ] )
[ ] ( [ ] ) ( [ ] )
( ) ( [ ] )
1 1 1 1
0 1 0 0 1 0 0 0 1 1 0
1 1 1 1
0 1 0 0 1 0 0 0 1 1 0
0
k k k k
k k k k
J r J r Y r
Y r Y r J r
λ λ γ λ λ
λ λ γ λ λ
+ −
− + =
(31)
(c) norms of eigenfunctions:
( ) ( ) ( ( ) ( ) )
( ) ( ) ( ) ( ) ( ) ( )
2 2 [1] 2 [1] 2 [1]
0 2 0 1 1 0 0 0 0 1 0 0
2 [1]
0
2 [1] [1] [1] [1] [1] [1]
0 1 1 0 1 1 0 0 0 0 0 0 0 1 0 0 1 0 0
2 1
2
R
k k k k
k
k k k k k k
N r J r Y r Y r
r J r Y r J r Y r J r Y r
λ λ λ
π λ
λ λ λ λ λ λ
= − + +
+ + −
( ) ( )
( ) ( )
2 2 [1] 2 [1] 2 [1]
0 0 0 0 1 0 0 1 1 0
1
2
k k kr J r λ J r λ Y r λ
− + (32)
B. Boundary conditions: G
i0 r
∂ =
∂ for r = r
i−1, G
i0 r
∂ =
∂ for r = r
i, i = 2, 3, ..., m − 1 (a) eigenfunctions:
[ ] ( [ ] ) ( [ ] ) ( [ ] ) ( [ ] )
0 0 0 1 0 0 0 1 0
i i i i i
k
J r
kY r
i kY r
kJ r
i kφ = λ λ − λ λ (33)
(b) eigenequation:
( [ ] ) ( [ ] ) ( [ ] ) ( [ ] )
1 1 0i 1 0i 1 1 0i 1 0i
0
i k i k i k i k
J r λ Y r λ Y r λ J r λ
−
−
−= (34)
(c) norms of eigenfunctions:
( ) ( ) ( ( ) ( ) )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( )
( ) ( )
2 2 [ ] 2 [ ] 2 [ ]
0 2 [ ] 2 1 1 0 0 1 0 1 1 0
0
2 [ ] [ ] [ ] [ ] [ ] [ ]
1 1 0 1 0 0 1 0 0 1 0 1 1 0 1 1 0
2 2 [ ] 2 [ ] 2 [ ]
1 0 1 0 1 1 0 1 0
2 1
2
1 2
R i i i
k i i k i k i k
i k
i i i i i i
i i k i k i k i k i k i k
i i i
i i k i k i k
N r J r Y r Y r
r J r Y r J r Y r J r Y r
r J r J r Y r
λ λ λ
π λ
λ λ λ λ λ λ
λ λ λ
− − −
− − − − −
− − −
= − + +
+ + −
− +
(35)
C. Boundary conditions: G
m0 r
∂ =
∂ for
1r = r
m−,
m0
m m
G
G
r γ
∂ − =
∂ for
r = r
m( α m
= α γ 1 ∞ ) (a) eigenfunctions:
[ ] ( [ ] ) ( [ ] ) ( [ ] ) ( [ ] )
0 0 0 1 1 0 0 0 1 1 0
m m m m m
k
J r
kY r
m kY r
kJ r
m kφ λ λ λ λ
− −
= − (36)
(b) eigenequation:
[ ] ( [ ] ) ( [ ] )
( ) ( [ ] )
[ ] ( [ ] ) ( [ ] )
( ) ( [ ] )
0 1 0 0 0 1 1 0
0 1 0 0 0 1 1 0
0
m m m m
k m k m m k m k
m m m m
k m k m m k m k
Y r Y r J r
J r J r Y r
λ λ γ λ λ
λ λ γ λ λ
−
−
− −
− − =
(37)
(c) norms of eigenfunctions:
( ) ( ) ( ( ) ( ) )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( )
( ) ( )
2 2 [ ] 2 [ ] 2 [ ]
0 2 1 1 0 0 1 0 1 1 0
2 [ ]
0
2 [ ] [ ] [ ] [ ] [ ] [ ]
1 1 0 1 0 0 1 0 0 1 0 1 1 0 1 1 0
2 2 [ ] 2 [ ] 2 [ ]
1 0 1 0 1 1 0 1 0
2 1
2
1 2
R m m m
k m m m k m k m k
k
m m m m m m
m m k m k m k m k m k m k
m m m
m m k m k m k
N r J r Y r Y r
r J r Y r J r Y r J r Y r
r J r J r Y r
λ λ λ
π λ
λ λ λ λ λ λ
λ λ λ
− − −
− − − − −
− − −
= − + +
+ +
− +
(38)
3. Numerical example
We determine the temperature in the two-layered cylinder by using properties of the Green’s functions which are derived in the previous section. The temperatures
T and
1T
2in the cylinder layers, submitted in the paper [7], are given in the form
( ) ( ) ( ) ( ) ( )
1 1 1 1 1 0 1 1 0
0
, , ; , , ; ,
t
T t r = k ∫ r y τ G t r τ r + r γ T
+τ G t r τ r d τ for r ∈ [ r r
0,
1] (39)
( ) ( ) ( )
1( ) ( ) [ ]
2 2 2 2 2 2 1 2 1 1 2
2 0
, , ; , , ; , for ,
t
T t r k r T G t r r r α y G t r r d r r r
γ τ τ τ τ τ
∞
α
= − ∈
∫
(40)
The function y τ ( ) is a solution of the Volterra integral equation of the first kind [7]:
( ) ( ) ( )
0
,
t
K t τ y τ d τ = F t
∫ (41)
where
( )
1 1 1(
1 1)
2 1 1 2(
1 1)
2
, , ; , , ; ,
K t k r G t r r k r α G t r r
τ τ τ
= + α
and
( )
2 2 2( )
2(
1 2)
1 0 1( ) (
1 1 0)
0
, ; , , ; ,
t
F t = ∫ k r γ T
∞τ G t r τ r − k r γ T
+τ G t r τ r d τ A numerical procedure to solution of the equation (42) is presented in paper [7].
The numerical calculations of the temperatures T and
1T in the cylinder are
2performed for the following geometrical and physical data: 0 . 04
0
=
r m, 0 . 045
1
=
r m,
06 .
2
= 0
r m, k
1= 7 . 5 ⋅ 10
−7m
2/s, k
2= 1 . 4 ⋅ 10
−6m
2/s, 0 . 08
1
=
α W (m K) ,
04 .
2
= 0
α W (m K) , α = 9 . 0
+
W (m 2 K) , α = 20 . 0
∞