0% found this document useful (0 votes)
18 views2 pages

On The Solution of The Heat Equation Wit

This document discusses the simulation of fin efficiency for various laminate samples and compares derived values with numerical results, showing deviations of less than 7 percent. It also presents a mathematical model for solving heat transfer problems with time-dependent coefficients, utilizing finite integral transforms. The findings indicate that the method can be effectively applied to problems with variable parameters.

Uploaded by

Ken Pepper
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)
18 views2 pages

On The Solution of The Heat Equation Wit

This document discusses the simulation of fin efficiency for various laminate samples and compares derived values with numerical results, showing deviations of less than 7 percent. It also presents a mathematical model for solving heat transfer problems with time-dependent coefficients, utilizing finite integral transforms. The findings indicate that the method can be effectively applied to problems with variable parameters.

Uploaded by

Ken Pepper
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/ 2

344 Shorter Communications

sequently quantified by calibration against Gardner’s curves. CONCLUSIONS

Table 1 shows these values of simulated I for various samples The accuracy obtainable with this analog is determined
of laminate. These values were then used in the simulation by the error in the calibrated value of 1. This can, with
of square and rectangular fins. Derived values of fin efficiency care. be kept within 10 per cent, and by varying the
are shown compared with numerical results in Fig. 3. number of paper thicknesses in the lamination, a wide range
of 1 may be simulated. A comparison of analytical and
numerical data with analog results shows a deviation of less
than k 7 per cent, and this technique has been used success-
fully in the evaluation of sheet fins which have not been
Table 1. Derived values of I for various samples oflaminate amenable to analytical or numerical solution.

No. of paper Derived I /I mean REFERENCES


Sample No. thicknesses (m-‘) (m-i)
1. D. Vitkovitch, Field Analysis, Chapters 1,2 and 5. D. Van
3 49.606 Nostrand, London (1966).
2 3 48.819
49.615 2. K. A. Gardner, Efficiency of extended surface, Trans. zyxwvutsrqp
Am.
Sot. M ech. Engrs 621 (1945).
3 3 49,600 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
J
3. T. E. Schmidt, Heat transfer calculations for extended
4 4 42.323
5 4 42.520 surface, Refrigng Engng 351- 357 (April 1949).
42.117 4. H. Zabronsky, Temperature distribution and efficiency of
6 4 43.307
7 4 42.520 a heat exchanger using square fins on round tubes, Trans.
Am. Sot. M ech. Engrs 22E, 119-122 (1955).
8 4 42,913
9 5 34.646 5. D. G. Rich, The efficiency and thermal resistance of
34.252 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHG
annular and rectangular fins, in Third International Heat
10 5 33.858
Transfer Conference, Vol. 3. p. 281 (1966).

zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
Inr. J. Heat M oss Transfrr. Vol. 18. pp. 344-345.
Pergamon Press 1975. Printed in Great Britain

ON THE SOLUTION OF THE HEAT EQUATION


WITH TIME DEPENDENT COEFFICIENT

M. D. MIKHAILOV
Applied Mathematics Centre, VMEI, Sofia, Bulgaria

(Received 30 June 1974)

NOMENCLATURE homogeneous region of arbitrary geometry V

A(N. T), B(N, z), cp(N, t), time dependent boundary aT(M, 4
functions defined on S; W+f, 4 ___aT = div [k(M , z) grad T(M , T)]
fo (M), initial distribution in V;
k(M , T), w(M , z), P(M , r), p(M , r), prescribed functions + CBW(M,4- ,0-f, T,]TM, 7)+P(M, 4,
defined in V; M EV, t>O (1)
M, point in V ;
N. point on 5; subject to the initial condition
n. outward normal of 5; 7(M, 0) = fo (M ) (2)
T(M , T), unsteady potential distribution defined in
equation (l), (2) and (3); and the boundary conditions
b(t), y(r). prescribed function defined in r; WN, 5)
I, time variable; AW, 7)____ + B(N, T)T(N, t) = q(N, T). (3)
$i (M , s), eigenfunctions ; an

/‘i(T), eigenvalues. In [l] is solved the particular case where w(M , T), k(M , T),
IN A KECENT paper [l], the author presented an analytical p(M , r), A(N, T) and B(N, r) are not functions of the time T.
solution for a large class of heat-transfer problems. The It is supposed that the solution of the problem can be
present communication complements [l] applying the represented in the form of an eigenfunction expansion, with
method of finite integral transforms for the solution of a the assumption that the eingenvalue problem
more general mathematical model of transfer process with
time and space dependent parameters. div CW, r)grad $1 (M, $1
Consider the following boundary value problem in a finite + [if (r)w(M, r) - o(M, r)]$i (M, r) = 0 (4)
Shorter Communications 345

subject to the boundary condition

Substituting
-
sT(M,T)~[w(M,r)~i(M.r)]dl’.

equation (7) in (11) and the result obtained


(11)

is granted for known. in (8), one gets


Equations (4H5) do not belong to the conventional
Sturm-Liouville family because the eigenfunctions Ii/i(M, T)
and eigenvalues pi(~) depend on zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
t. The idea of time
dependent eigenvalues we borrow from Vidin, who presented
in [2] an approximate method of temperature field
calculations in an infinite cylinder that the convective heat- =‘fi i=l,2,... (12)
transfer coefficient is an arbitrary time function. y(T)
To solve equation (1) at the conditions (2) and (3), the
finite integral transform
-
is to be used.
z (T) =
s
Y
w(M,T)$i(M, T)T(M, T)dV (6) Qji(T)=
s ”
~j(M,T)~[w(M,T)~i(M,T)]dV.
The functions T(T) being obtained from (12) the desired
(13)

It follows, from the orthogonality of eigenfunctions function T(M, T) straight forward determined by the
tii(M, T), that T(M, T) can be expanded into a series as inversion formula (7).
If w(hf,~), k(hf,~), zyxwvutsrqponmlkjihgfedcbaZYX
~(M ,T), A(N,r) and B(N,r) were

T(M, T) = f
'h(M,T)
%(T) (7)
independent zyxwvutsrqponmlkjihgfedcba
of time, i.e. I(li(M, T) and pi (T) were independent

i=1 of time too, the system (12) would have been a system of
w(M, T)l,@(hf,T) zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA
dV uncoupled ordinary, first order, linear differential equations.
s ”
which are solved in [l].
Some mathematical operations fully analogous to these in The method presented here can easily be applied to
[ 1] leads to : problems analogous to the ones treated in [3,4] but with
time dependent coefficients. In [S] are presented the solu-

s
y(T) MM, +ki (M>5) 7

aWf> 7Jdv tion of the particular case when Y(T) = 1, /I(T) = 0 and
w(M, r), k(M, T), p(M, T), A(N, T) are time independent.
+ bf(d - p(T)]T (T)= gi (T) (8) the same paper a method is described for the solution of
In

the system (12) through s-order approximation.


where

iNT)- [@i(N, d/an]


Si (7)=
s s
k(N, r)cp(N, T) +i

A@, 5)+&N, T)
dS

REFEREN CES

1. M. D. Mikhailov, General solution of the heat equation


in finite regions, Int. J. Engng Sci. 10, 577-591 (1972).
+ y$i(M,W(M~~)dV. (9)
s 2. Yu. V. Vidin, Heating of a cylinder with an internal heat
From the relation source and variable heat-transfer coefficient, J. Engng
Phys. 11, 166-170 (1966).

; [w(M, @i (M, T)Tb+‘f,T,] = MM, T)$i (M, T) ‘7 3. M. D. Mikhailov, General solutions of the coupled
diffusion equations, Znt. J. Engng Sci 11, 235-241 (1973).
4. M. D. Mikhailov, General solutions of the diffusion
+WKT)~[W(M,T)~,(M,T)] (10) equations coupled at boundary conditions, Int. J. Heat
Mass Transfir 16,2155-2164 (1973).
follows 5. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM
M. N. &i+k and R. L. Murray, On the solution of linear
diffusion problems with variable boundary condition
parameters, Trans. Am. Sot. Me&. Engrs C, 96, 48-51
(1974).

You might also like