Since large power transformers belong to the most valuable assets in electrical power networks it is
suitable to pay higher attention to these operating resources. Thermal impact leads not only to long-term
oil/paper-insulation degradation; it is also a limiting factor for the transformer operation. Therefore, the
knowledge of the temperature, especially the hottest spot (HST) temperature, is of high interest. This paper
presents steady state temperature distribution of a power transformer layer-type winding using conjugated
heat transfer analysis, therefore energy and Navier-Stokes equations are solved using finite difference
method. Meanwhile, the effects of load conditions and type of oil on HST are investigated using the model.
Oil in the transformer is assumed nearly incompressible and oil parameters such as thermal conductivity,
special heat, viscosity, and density vary with temperature. Comparing the results with those obtained from
finite integral transform checks the validity and accuracy of the proposed method
Optimization of a Shell and Tube Condenser using Numerical Method
Report
Share
1 of 11
Download to read offline
1
2
3
4
5
6
7
8
9
10
11
More Related Content
Layer-Type Power Transformer Thermal Analysis Considering Effective Parameters On The Temperature Rise
1. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
1
Layer-Type Power Transformer Thermal Analysis
Considering Effective Parameters On The
Temperature Rise
M.Ghanbaria
, M.A.Taghikhanib
a
Department of Electrical and Computer Engineering, Saveh Branch, Islamic Azad
University, Saveh, Iran
b
Department of Engineering, Imam Khomeini International University, Qazvin, Iran
ABSTRACT
Since large power transformers belong to the most valuable assets in electrical power networks it is
suitable to pay higher attention to these operating resources. Thermal impact leads not only to long-term
oil/paper-insulation degradation; it is also a limiting factor for the transformer operation. Therefore, the
knowledge of the temperature, especially the hottest spot (HST) temperature, is of high interest. This paper
presents steady state temperature distribution of a power transformer layer-type winding using conjugated
heat transfer analysis, therefore energy and Navier-Stokes equations are solved using finite difference
method. Meanwhile, the effects of load conditions and type of oil on HST are investigated using the model.
Oil in the transformer is assumed nearly incompressible and oil parameters such as thermal conductivity,
special heat, viscosity, and density vary with temperature. Comparing the results with those obtained from
finite integral transform checks the validity and accuracy of the proposed method.
KEYWORDS
Power transformer; Layer-Type Winding; Thermal Analysis; Conjugated Heat Transfer; Finite difference
method.
1. INTRODUCTION
In a power transformer a part of the electrical energy is converted into the heat. Although this part
is quite small comparing to total electric power transferred through a transformer, it causes
significant temperature rise, which represents the limiting criteria for possible power transfer
through a transformer. That is why the precise calculation of temperatures in critical points (top
oil and the hottest solid insulation spot) is of practical interest. Thermal impact leads not only to
long-term oil/paper-insulation degradation; it is also a limiting factor for the transformer
operation [1], [2]. Therefore, the knowledge of the temperature, especially the hottest spot
temperature, is of high interest. If the temperature rise goes beyond the permissible value, the
load of transformer must be reduced or an auxiliary transformer is used in order to preserve the
insulation from deterioration. For an oil-immersed transformer, the oil surrounds the transformer
body. Oil is a nearly incompressible fluid and density changes due to temperature rise, therefore
oil moves in the transformer. The heat transferred by convection is the most important method of
heat transfer.
2. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
2
Hottest spot temperature must not exceed the prescribed value in order to avoid insulation faults.
Furthermore, an accurate computation of the HST helps in a realistic estimation of the reliability
and remaining life of the transformer winding insulation. The classical approach has been to
consider the hottest-spot temperature as the sum of the ambient temperature, the top-oil
temperature rise, and the hottest-spot-to-top-oil temperature gradient. A hottest spot temperature
calculation is given in the International Standards [3]-[5]. The algorithm for calculating the
hottest spot temperature of a directly loaded transformer using data obtained in a short circuit
heating test is given in [6]. Heat transfer theory results from winding to oil are exposed in [7], [8].
The structure of a transformer winding is complex and does not conform to any known geometry
in the strict sense. Under general conditions, the transformer windings can be assumed
cylindrical; hence, a layer or a disc winding is a finite annular cylinder. The thermal and physical
properties of the system would be equivalent to a composite system of insulation and conductor.
In this paper, author has proposed a procedure for obtaining the temperature distribution in the
power transformer and the effects of load conditions and type of oil are investigated using the
model. For this reason energy and Navier-Stokes equations are solved using finite difference
method. Therefore, a code has been provided under MATLAB software. The model can be used
for temperature calculation on the arbitrary change of current and outside air temperature. In the
paper, thermal model, energy, and Navier-Stokes equations are given in section 2. Results and
discussion of the proposed work have been provided in section 3.
2. THERMAL MODEL
Energy equation for Newtonian incompressible fluid (or nearly incompressible fluid) such as oil
in two dimensions is [9]
)
(
)
(
)
)
(
)
(
(
y
T
k
y
x
T
k
x
y
T
C
V
x
T
C
V
p
y
p
x
∂
∂
⋅
∂
∂
+
∂
∂
⋅
∂
∂
=
∂
⋅
∂
+
∂
⋅
∂
⋅
(1)
Where T is temperature, Vx is velocity in x direction, Vy is velocity in y direction, k is thermal
conductivity, is density and Cp is special heat of the oil in the transformer. In equation (1), oil
properties vary with temperature [1], [10], [11]. The temperature dependence of oil properties are
given as follows:
)
273
exp( 2
1
+
⋅
=
T
a
a
(2)
T
a
a
C
p ⋅
+
= 4
3 (3)
T
a
a ⋅
+
= 6
5
(4)
T
a
a
k ⋅
+
= 8
7 (5)
9
a
=
(6)
Where : is viscosity and is volumetric expansion coefficient. The nine constants for
transformer oil and silicon oil have been listed in Table 1 [3, 4]. It is generally valid for all types
of transformer oil that the variation of the oil viscosity with temperature is much higher than the
variation of other oil properties [11]. Thus, all oil physical properties except the viscosity can be
3. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
3
replaced by a constant. However, the variations of all oil properties with temperature have been
considered in this paper.
Table1. Oil constants
Oil constant Transformer oil Silicon oil
a1 0.0000013573 0.00012127
a2 2797.3 1782.3
a3 1960 1424
a4 4.005 2.513
a5 887 989
a6 -0.659 -0.870
a7 0.124 0.138
a8 -0.0001525 -0.00009621
a9 0.00086 0.00095
In (1) velocity is unknown, and then we must solve Navier-Stokes equations. Navier-Stokes
equations in two dimensions for Newtonian incompressible fluid are [9]
Continuity Equation:
0
=
∂
∂
+
∂
∂
y
V
x
V y
x
(7)
Momentum Equations:
))
(
)
2
(
(
)
.(
x
V
y
V
y
x
V
x
x
P
y
V
V
x
V
V
y
x
x
x
y
x
x
∂
∂
⋅
+
∂
∂
⋅
∂
∂
+
∂
∂
⋅
⋅
∂
∂
+
∂
∂
−
=
∂
∂
+
∂
∂
(8)
)
(
))
2
(
)
(
(
)
.(
0 ref
y
y
x
x
y
x
x
T
T
g
y
V
y
x
V
y
V
x
y
P
y
V
V
x
V
V
−
⋅
⋅
⋅
+
∂
∂
⋅
⋅
∂
∂
+
∂
∂
⋅
+
∂
∂
⋅
∂
∂
+
∂
∂
−
=
∂
∂
+
∂
∂
(9)
Where P is pressure and g is gravitational acceleration .The heat conduction equation for core and
windings is written [10], [11]
0
2
2
2
2
=
+
∂
∂
⋅
+
∂
∂
⋅ Q
y
T
k
x
T
k y
x (10)
Where kx and ky are thermal conductivity in x and y directions respectively. The term Q is the
volumetric heat source function and has been modified here to take care of variation of electrical
resistance of copper with temperature. The heat source term Q can be of the form
)]
(
1
[
0 ref
c T
T
Q
Q −
⋅
+
⋅
= (11)
4. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
4
Where c is the temperature coefficient of electrical resistance of copper wire. With this
representation, the function Q becomes temperature dependent, distributed heat source. Thermal
conductivities are unequal in different directions. Thermal conductivity has been treated as a
vector quantity, having components in both radial and axial direction. Resultant thermal
conductivity of the system is [10]
2
2
y
x k
k
K +
= (12)
Where
)
log
...
log
log
(
log
1
2
2
3
1
1
2
1
n
n
n
n
x
k
r
r
k
r
r
k
r
r
r
r
k
−
+
+
+
=
in
cu
cu
in
in
cu
in
cu
y
k
t
k
t
t
t
k
k
k
⋅
+
⋅
+
⋅
⋅
=
)
(
Term K represents resultant thermal conductivity of insulation and conductor system. In this
paper, the bottom oil temperature rise over ambient temperature has been calculated as
n
l
l
r
fl
u
R
R
I
)
1
1
(
2
+
+
⋅
=
(13)
Where u is the bottom oil temperature rise over ambient temperature, fl is the full load bottom
oil temperature rise over ambient temperature obtained from an off-line test, Rl is the ratio of load
loss at rated load to no-load loss. The variable Ir is the ratio of the specified load to rated load.
rated
r
I
I
I = (14)
The exponent n depends upon the cooling state. The loading guide recommends the use of n=0.8
for natural convection and n=0.9-1.0 for forced cooling.
3. RESULTS AND DISCUSSION
A code has been provided for finite difference solution of energy and Navier-Stokes equations
using MATLAB 7.1. Figure 1 shows cross section of a 50KVA, 20KV/400V power transformer
in two dimensions. Dimensions and specifications of the power transformer have been
summarized in Table 2. Table 3 shows the power transformer losses.
5. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
5
Figure1. Three phase power transformer in two dimensions
Table2. Dimensions and specifications of the proposed power transformer
Core diameter (cm) 10
Width of each window (cm) 12
Height of window (cm) 28
Thickness of LV winding (cm) 1.2
Thickness of HV winding (cm) 2
Rated power 50KVA
HV voltage 20kV
LV voltage 400V
HV current 1.44A
LV current 72A
Table3. Power transformer losses
Losses (w) Value
Core 158
DC of LV windings 384
DC of HV windings 534
Eddy currents of LV windings 64
Eddy currents of HV windings -
We assume that inlet oil temperature is the temperature base. Therefore, we have
(15)
100
,
,
×
−
=
∆
oil
b
oil
b
T
T
T
T
6. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
6
Where Tb,oil is the bottom oil temperature. Figure 2 shows temperature distribution from LV
winding (axis symmetry in height direction) in one per unit (p.u.) load with oil natural cooling
(ON). It can be pointed out that for LV winding, maximum temperature location is around 80%
of winding height from the bottom and at about 50% of radial thickness of the layer. Temperature
distribution from HV winding has been shown in figure 3. It can be observed that the maximum
temperature occurs in the neighborhood of 55% of the axial and 50% of the radial thickness of the
layer.
0 0.05 0.1 0.15 0.2
32
34
36
38
40
42
Height of LV winding(m)
HST
rise
over
bottom
oil
temperature(%)
Figure2. Temperature distribution in height direction of LV winding with transformer oil natural cooling
7. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
7
0 0.05 0.1 0.15 0.2
44
46
48
50
52
54
56
58
60
62
64
Height of HV winding(m)
HST
rise
over
bottom
oil
temperature(%)
Figure3. Temperature distribution in height direction of HV winding with transformer oil natural cooling
Figure 4 shows HST rise over bottom oil temperature versus load for two types of oils and
figure 5 shows HST versus load at Tamb=25(°C).
0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
40
50
60
70
80
90
100
110
120
Load(p.u.)
HST
rise
over
bottom
oil
temperature(%)
Transformer oil
Silicon oil
Figure4. HST rise over bottom oil temperature versus load at Tamb=25(°C) with oil natural cooling
8. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
8
0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
60
70
80
90
100
110
120
130
140
150
Load(p.u.)
HST(°C) Transformer oil
Silicon oil
Figure5. HST versus load at Tamb=25(°C) with oil natural cooling
It can be pointed out from figures 4 and 5 that the HST value is low when the transformer oil is
used for cooling, therefore the transformer oil is better than silicon oil in the cooling of the power
transformer when load changes. Table 4 shows comparison of the proposed method with finite
integral transform used in [10] for power transformer oil. HST locations at different loading are
given in Table 5.
Table4. HST (°C) magnitudes at Tamb=25(°C)
Load (p.u.) Proposed Analytical [10]
0.6 51.72 53
0.7 58.35 56
0.8 65.71 66
0.9 73.78 75
1.0 82.54 80
1.1 91.99 93
1.2 102.15 101
1.3 113.01 115
1.4 124.59 125
1.5 136.91 139
9. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
9
Table5. HST locations at Tamb=25(°C)
Load (p.u.)
Proposed (radial thickness%-
height%)
Analytical [10] (radial thickness%-
height%)
0.6 50 / 54.17 44 / 52
0.7 50 / 54.17 45 / 53.5
0.8 50 / 54.58 46 / 54
0.9 50 / 54.58 48 / 54.5
1.0 50 / 55 50 / 55
1.1 50 / 55 50 / 55
1.2 50 / 55.42 50 / 55
1.3 50 / 55.42 50 / 55
1.4 50 / 55.42 50 / 55.5
1.5 50 / 55.42 50 / 55.5
Reference [10] has proposed an analytical method based on boundary value problem of heat
conduction in power transformer winding using finite integral transform techniques. This
technique uses Fourier transform and Hankel transform and finds temperature based on
polynomials with infinite order (sequential series) of space variables and time, and then
approximates temperature with finite order. Analytical method in [10] is suitable for cylindrical
winding with rectangle cross-section, but this limitation is not for finite difference method. In the
other hand, if we assume that we can increase order of polynomial, solution time will rise and
analytical method will not differ with numerical method.
4. CONCLUSION
A numerical study was conducted to investigate temperature distribution in layer-type power
transformer for a variety of load conditions and type of oil. On the other hand, an attempt has
been made to suggest a method to improve the accuracy of prediction of the temperature of the
hottest spot in power transformer. Heat transfer partial differential equations were solved
numerically with finite difference method. The heat source function has been taken as
temperature dependent and has been directly incorporated in the heat conduction equation for
core and windings. The purely numerical approach followed in this paper seems to correspond
reasonably well with results of analytical calculations [10]. In addition, it can be pointed out that
transformer oil is better than silicon oil in cooling of the power transformer when load changes.
NOMENCLATURE
Cp : specific heat of oil (J.kg-1
.K-1
)
g: gravitational acceleration (m.s-2
)
Ir: ratio of the specified load to rated load
k: thermal conductivity of oil (W.m-1
.K-1
)
K: equivalent thermal conductivity (W.m-1
.K-1
)
kcu : thermal conductivity of copper (W.m-1
.K-1
)
kin : thermal conductivity of insulation (W.m-1
.K-1
)
kx : thermal conductivity in x direction (W.m-1
.K-1
)
ky : thermal conductivity in y direction (W.m-1
.K-1
)
10. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
10
n : temperature rise exponent due to bottom oil
P: pressure (N.m-2
)
Q : volumetric heat source function (W.m-3
)
Q0: reference volumetric heat source (W.m-3
)
ri’s : radius of insulation and conductor layers (m)
Rl : loss ratio = load loss/no load loss
T: temperature (K)
Tref : reference temperature (K)
Tb,oi : bottom oil temperature (K)
tcu: thickness of conductor (m)
tin : thickness of insulation (m)
Vx: velocity in x direction (m.s-1
)
Vy: velocity in y direction (m.s-1
)
Abbreviations
HST: hottest spot temperature
HV: high voltage
LV: low voltage
ON: oil natural cooling
p.u.: per unit
Greek symbols
: coefficient of volumetric expansion of oil (K-1
)
u : bottom oil temperature rise over ambient temperature (K)
fl: bottom oil temperature rise over ambient temperature at rated load (K)
: density of oil (kg.m-3
)
0: reference density of oil (kg.m-3
)
c : temperature coefficient of electrical resistance (K-1
)
: viscosity of oil (kg.m-1
.s-1
)
REFERENCES
[1] L. W. Pierce, (1992) “An investigation of the thermal performance of an oil filled transformer
winding,” IEEE Transactions on Power Delivery, 7(3), pp. 1347-1358.
[2] L. W. Pierce, (1994) “Predicting liquid filled transformer loading capability,” IEEE Transactions on
Industry Applications, 30(1), pp. 170–178.
[3] IEC Standard, IEC60076-7 (2006) "Loading guide for oil immersed transformers".
[4] IEEE Standard, C57.91-1995 (1996) "IEEE guide for loading mineral oil immersed transformer".
[5] IEEE Standard, 1538-2000 (2000) "IEEE guide for determination of maximum winding temperature
rise in liquid-filled transformers".
[6] Z. Radakovic, K. Feser, (2003) “A new method for the calculation of the hot-spot temperature in
power transformers with ONAN cooling”, IEEE Transactions on Power Delivery, 18(4), pp. 1–9.
[7] G. Swift, T. Molinski, W. Lehn, R. Bray, (2001) “A fundamental approach to transformer thermal
modeling—Part I: Theory and equivalent circuit”, IEEE Transactions on Power Delivery, 16(2), pp.
171-175.
[8] W. H. Tang, O. H. Wu ,Z. J. Richardson, (2002) “Equivalent heat circuit based power transformer
thermal model”, IEE Electric Power Application, 149(2), pp. 87-92.
11. Advanced Energy: An International Journal (AEIJ), Vol. 1, No. 1, January 2014
11
[9] F. P. Incropera, D. P. DeWitt, (1996) “Fundamentals of heat and mass transfer”, 4th edition, New
York/USA: J. Wiley & Sons.
[10] M. K. Pradhan, T. S. Ramu, (2003) “Prediction of hottest spot temperature (HST) in power and
station transformers”, IEEE Transactions on Power Delivery, 18(4), pp. 1275–1283.
[11] M. A. Taghikhani, A. Gholami, (2008) “Temperature Distribution in Power Transformer Windings
with NDOF and DOF Cooling”, PES General Meeting, Pittsburgh, PA, USA, 20-24 July .