Simulation and Optimization of Cooling Tubes of Transformer for Efficient Heat Transfer

International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-3, Issue-2, Feb- 2017]
https://dx.doi.org/10.24001/ijaems.3.2.23 ISSN : 2454-1311
www.ijaems.com Page | 141
Simulation and Optimization of Cooling Tubes of
Transformer for Efficient Heat Transfer
Jackwin Vincent K, Dr V.V.Prathibha Bharathi
1
Department of mechanical Engineering, Malla Reddy college of Engineering, Secunderabad, India
Abstract— Temperature variation with in the transformer
affects the life and efficiency of the distribution
transformer. The top oil temperature in the transformer
depends on the type of cooling and cooling ducts/fins
design and their layout. The present project investigates
methods of onan transformer cooling system by means of
increasing heat transfer rate by implanting the axial
groves along with fins and porous region within in the
cooling tubes and further optimization of the cooling
process by adjusting the gravity by orienting the tube.
This study is carried out by means of numerical analysis
by simulating Transformer geometry in Ansys Fluent
.Real case geometry of distribution transformer is used in
this simulation
Keywords— Ansys Fluent, CFD, Distribution
transformer cooling, numerical Analysis, Optimization
I. INTRODUCTION
There are several factors contributing to power losses in
transformers. These are copper losses, which represent the
major source of losses in a transformer, and core losses;
namely hysteresis and eddy current losses. These losses
components are produced in the form of heat energy
which should be dissipated in a quick and rather efficient
manner. If the transformer has otherwise failed to get rid
of such heat generated, many problems could arise and in
some cases severe consequences may occur.
In fact, the improperly dissipated heat would further
accumulate and thus cause the transformer temperature to
increase. This process may lead to failure of paper
insulation and liquid insulation medium of the
transformer. Furthermore, excessive heat may result in
damage of the transformer windings, the matter which, in
particular, is considered as a catastrophe for expensive
high power rating transformers. Therefore, numerous
ways are introduced to keep the temperature within
acceptable limits which in turn would help to maintain a
long lifetime of the transformer. One such way is onan
cooling of transformer which means oil natural air natural
which is in distribution transformer. This method makes
use of natural convection of transformer oil and air.
The present project investigates methods of transformer
cooling system by means of increasing heat transfer rate
by implanting the axial groves along with fins and porous
region within in the cooling tubes. And further
optimization of the cooling process by adjusting the
gravity by orienting the tube. Five optimization cases is
studied This study is carried out by means of numerical
analysis by simulating Transformer geometry in Ansys
Fluent.
An experimental investigation of Temperature
distribution in a distribution transformer and Experiment
investigation of free convection heat transfer from an
array of vertical tube is carried out by
A.Satyanarayana.Reddy, Suresh Akella, and AMK.
Prasad [1][2].Thermal modeling of natural convection
cooled oil immersed distribution transformer is carried
out by Robert Michael Marko[3]. Transformer cooling
system improvement has been extensively studied
through analytical or advanced numerical techniques by
Miguel E. Rosillo[5]. Eleftherios I. Amoiralis and Stefan
D.L. ŢĂLU [4] developed a method for optimum
transformer tank panels geometry for the cooling system.
Y. X. Wang, H. B. Ma and G. P. Peterson [6] did
Investigation of the Temperature Distribution on Radiator
Fins with Micro Heat Pipes.
S.A.Nada and M.Mowad [7] performed an experimental
study on free convection from a vertical and inclined
semicircular cylinder at different orientations at constant
heat flux. The experiments were carried at four inclination
angles (0˚, 30˚, 45˚ and 60˚) of the semicircular cylinder.
The results showed that the average Nusselt number
increases as the inclination angle of the semicircular
cylinder increases. Alegi, g. and Black simulated Real-
time thermal mode1 for an oil immersed forced-air cooled
transformer. They contributed the study of forced
convection in the area of distribution transformer.
II. IMPROVEMENT OF TRANSFORMER
COOLING SYSTEM
The transformer consists of a closed magnetic circuit (the
core) with two coils, or windings, of insulated conductors
wound around the core. The current in the input, or
primary, coil creates a magnetic field. This field induces a
voltage across the output, or secondary coil. The
secondary coil of a distribution transformer is designed to
deliver energy at a lower level of voltage relative to the
input voltage. For a distribution transformer, the high

voltage winding is the primary coil and the low voltage
winding is the secondary coil.
As the current passes through the copper or Aluminum
windings, energy is lost because of electrical resistance
power losses (Power = current2 x Resistance). The energy
losses raise the temperature of the windings. Various
methods of dissipating energy to the surroundings are
employed to cool the windings. The temperature rise is
controlled by transferring energy generated from the
transformer to a cooling fluid, such as oil or air. Excess
heating accelerates the aging of the transformer
insulation.
When the oil paper insulation is subjected to thermal
stress, the oil and paper components of the insulation may
be irreversibly damaged. The thermal stress reduces the
mechanical and dielectric performance of the insulation
.There is a need to optimize the design and application of
transformers. Manufacturers and utilities try to reduce the
capital and operational costs of transformers. The main
objectives of this Project are:
 Numerical simulation of natural convection
process in a onan distribution transformer and find
out maximum hot spot temperature at the core and
windings
 To carry out design optimization in cooling tubes
to increase heat transfer and to reduce hot spot
temperature of core.
 Comparative study of five optimization cases for
efficient heat transfer
 Case 1: normal case without any modification
 Case 2: Implanting axial grooves with fins in
cooling tubes
 Case 3: With porous media inside cooling tubes
 Case 4: With porous media and axial grooves with
fins inside tubes.
 Case 5: With inclination of cooling tubes
III. NUMERICAL SIMULATION AND
OPTIMIZATION
The fluid flow and heat transfer in a transformer is
modeled in this work using the Navier-Stokes equations
for incompressible flow, and the conservation of energy
equation. The mathematical mode1 used in the thesis
consists of four equations: energy conservation,
momentum conservation, and mass conservation. The
solution of the energy equation produces the temperature
field, and the momentum equations determine the
velocities in the fluid. The mass conservation equation
ensures that fluid entering a control volume also leaves
the control volume and is used as the constraint on
pressure. The energy conservation equation is given by
Equation (1). The first term on the left hand side of the
energy conservation equation represents the energy
storage. The second and third terms represent the
advection of energy by the moving fluid. On the right
hand side of the equation, the first three terms are the
energy diffusion term and the last term represents the
energy generation per unit volume.
‘Cp ∂)/∂t (ρT) +Cp ∂/∂x (ρUT)+Cp ∂/∂y (ρVT)+Cp ∂/∂z
(ρWT)=∂/∂x Kx ∂T/∂x+∂/∂y Ky ∂T/∂y+∂/∂z Kz ∂T/∂z+Q'
(1)
The momentum conservation equations in three
dimensions, Equations (2) (3), (4) are used to calculate
the velocity fields. The first term in each equation
represents the change in momentum over the time. The
second, third and fourth terms on the left hand side of the
equations represent the advection of momentum. The first
three terms on the right-hand side of the equations
account for the net viscous forces, and the fourth term
represents the pressure forces. The last term of the y
direction momentum equation represents the buoyancy
force used when solving a natural convection problem.
∂/∂t (ρU) +∂/∂x (ρUU) +∂/∂y (ρVU)+∂/∂z (ρWU)=∂/∂x μ
∂U/∂x+∂/∂y μ ∂U/∂y+∂/∂z μ ∂U/∂z-∂P/∂x (2)
∂/∂t (ρV) +∂/∂x (ρUV) +∂/∂y (ρVV)+∂/∂z (ρWV)=∂/∂x μ
∂V/∂x+∂/∂y μ ∂V/∂y+∂/∂z μ ∂W/∂z-∂P/∂y+ρ∞βg(T-T∞)
(3)
∂/∂t (ρW)+∂/∂x (ρUW)+∂/∂y (ρVW)+∂/∂z (ρWW)=∂/∂x
μ ∂W/∂x+∂/∂y μ ∂W/∂y+∂/∂z μ ∂W/∂z-∂P/∂z (4)
Natural convection is the body force driving fluid flow
within a naturally oil cooled transformer. A buoyancy
force is the net effect of body forces acting on a fluid in
which there are density gradients. For this model, the
density gradient is induced by a temperature gradient in
the fluid, and the body force is due to the gravitational
field, g, acting in the negative x direction.
The buoyancy force in the momentum equation is
modeled by using the Boussinesq approximation .In that
term, ρ_∞, is the average fluid density, β is the thermal
coefficient of volume expansion, and T_∞, is the average
fluid temperature. Equation (5) is the continuity equation.
In that equation the first term accounts for the mass
storage, and the other three terms describe the change of
mass flow in both x, y and z directions, respectively.
( ∂)/∂t (ρ) +∂/∂x (ρU) +∂/∂y (ρV) +∂/∂z (ρW) =0 (5)
The mathematical model consists of a set of differential
equations expressing conservation of mass, momentum,
and energy, over a Cartesian domain. The fluid was
assumed to have Newtonian and laminar –Turbulent flow
characteristics. The energy equation was modified to
allow for conjugate heat transfer between a solid and the
surrounding fluid. A method for modeling the windings
as a homogeneous material was also introduced. The
transport equations were discretized using a finite volume

approach, and the pressure-velocity coupling was handled
using the SIMPLE algorithm.
3.1 CFD analysis using Ansys Fluent
Fig 1: General process of CFD analysis
3.1.1Preprocessing & Solver set up
Geometry considered for analysis is that of an ONAN
distribution transformer (oil natural air natural).Model
contains a source, oil tank, cooling transformer oil, oil
collecting region and cooling tubes or radiators. Source
includes copper windings and coil. Major heat is
dissipated from the source due to eddy current loss and
copper loss and heat transfer occur due to convection,
conduction and radiation.
As Transformer is oil immersed to provide cooling by
natural convection, major heat transfer is occurred by
convection, so heat transfer by conduction and radiation is
negligible and neglected from Analysis. Geometry is
created using CatiaV5 R20.
Meshing is done in ANSYS ICEM CFD, with minimum
skewness of 0.82 .Maximum face size is 6mm and
maximum size is 12mm. Minimum edge length is 2mm.
Meshing is defined as discretizing the cad model into
finite element model. Simulation is done in Ansys
Fluent .Before 2006 Fluent is a separate solver, now
Ansys acquired Fluent and is within the Ansys workbench
module .Open the Setup option of Ansys fluent project
• Select Solver setup > General > among solver settings
choose the pressure based solver. CFD fluent consist of
two solvers for solving the CFD process i.e. pressure
based and density based. Here we choose pressure based;
this is because the fluid is incompressible. Choose the
velocity formulation as absolute and time option as steady
state. This is because flow is considered to have constant
properties with respect to time.
Select Solver setup > Models > Choose energy
equation on.. To solve the heat transfer problems the basic
energy equation has to be used and this can be used by
switching on the energy equation. Transition K-KL-
omega model is one of the most commonly used
turbulence models to model Transition flow. It is a three
equation model that means, it includes three extra
transport equations to represent the turbulent properties of
flow. The K-KL-omega transition model is used to
predict boundary layer development and calculate
transition onset. This model can be used to effectively
address the transition of the boundary layer from a
laminar to a turbulent regime. The K - KL -omega
model is considered to be a three-equation eddy-viscosity
type, which includes transport equations for turbulent
kinetic energy (K_T), laminar kinetic energy (K_L), and
the inverse turbulent time scale (ω).
Fig 2: Catia model used in the analysis

Fig 3: Meshed model
Select Cell zone conditions and choose the appropriate
materials for the zones. Transformer oil is defined to fluid
zone. Copper is assigned to part-source and in source
terms Energy source of 1302082 w/m^3 is defined to
represent energy generation in core. In part- porous
domain switch on the porous zone to model
corresponding case if not uncheck the option. In the
solution method choose pressure velocity coupling
scheme as SIMPLE.
SIMPLE is the acronym of semi implicit method for
pressure linked equations. It is the default scheme and the
robust one which can be used for all types of analysis.
Choose the gradient option as ‘least square cell based’. It
is the default method; has the good accuracy and the same
time it is computationally less intensive. Choose body
force weighted option for the pressure. It is used for
gravity has prominent role. Choose momentum, turbulent
kinetic energy and laminar kinetic energy and specific
dissipation rate and energy as first order upwind.
Select the monitors > residuals. Click edit. In the dialogue
box that appears edit the convergence criteria of
continuity, x velocity, y velocity, z velocity, k, Omega as
0.001. Energy as 0.000001 Close the dialogue box.
In the solution initialization select the hybrid initialization
and click initialize. By starting CFD simulation we must
provide fluent with an initial guess for the solution flow
field. Hybrid initialization is a collection of recipes and
boundary interpolation methods. It solves Laplace's
equation to determine the velocity and pressure fields. All
other variables, such as temperature, turbulence, species
fractions, volume fractions, etc., will be automatically
patched based on domain averaged values or a particular
interpolation recipe. In the Run calculation window edit
the number of iterations as 10000 and click calculate.
Within 500 iterations solver will give the converged
results with required accuracy.
Table.1: Material properties used
Table 2: Boundary conditions

3.1.2 Postprocessing
Fig.4: Geometry of different cases used in the analysis
IV. RESULTS AND OBSERVATIONS
Table.3: Values of output variables for all cases
4.1 Temperature contours
Fig: 5: Temperature contour for all 5 cases
4.2 Graph comparing Results
Fig.6.Graph showing max temperature vs all cases
Fig.7: Graph showing turbulence kinetic energy vs all
cases

Fig.8: Graph comparing different velocities
V. CONCLUSION
Numerical simulation of free convection in onan
distribution transformer is done and a mixed laminar and
turbulent flow is found to be occurred, also design
modification in cooling tubes provides better turbulent
generation and increased heat transfer. Cooling tube with
axial grooves and fins are found to be most effective by
decreasing hot spot temperature by 34 degree Celsius
.Also from the simulation it is found that natural
convection flow of transfer oil due to density difference
with increase in temperature can be effectively simulated
using computational fluid dynamics. .
Fig.9: Graph comparing Turbulence eddy frequency vs
cases
REFERENCES
[1] International Journal of Emerging Technology and
Advanced Engineering ISSN 2250-2459, ISO
9001:2008 Certified Journal, Volume 4, Issue 3,
March 2014) 253Experimental Study of Free
Convection Heat Transfer From Array Of Vertical
Tubes At Different Inclinations by
A.Satyanarayana.Reddy, Suresh Akella, AMK.
Prasad
[2] Vol. 3 Issue 4, April – 2014 International Journal of
Engineering Research & Technology (IJERT) ISSN:
2278-0181 Investigation of the Temperature
Variation in Distribution Transformer Cooling
System by A.Satyanarayana.Reddy, Suresh Akella,
AMK. Prasad
[3] Thermal Modeling of a Natural-Convection-Cooled,
Oil-Immersed Distribution Transformer by Robert
Michael Marko.
[4] Ştefan D.L. ŢĂLU1, Mihai D.L. ŢĂLU2-
Dimensional optimization of frontal radiators Of
cooling system for power transformer 630 Kva
20/0.4 kv in terms of maximum heat transfer, U.P.B.
Sci. Bull. Series C, Vol. 72, Iss. 4, 2010
[5] Miguel E. Rosillo, Carlos A. Herrera, and Guillermo
Jaramillo, IEEE Transactions On Power Delivery,
Vol. 27, No. 4, October 2012
[6] Y. X. Wang, H. B. Ma and G. P. Peterson
Investigation of the Temperature Distribution on
Radiator Fins with Micro Heat Pipes, Journal Of
Thermo physics And Heat Transfer ,Vol. 15, No. 1,
January–March 2001
[7] S. A. Nada a and M. Mowad, Free convection from
a vertical and inclined semicircular cylinder.

Simulation and Optimization of Cooling Tubes of Transformer for Efficient Heat Transfer

More Related Content

What's hot (18)

Viewers also liked (9)

Similar to Simulation and Optimization of Cooling Tubes of Transformer for Efficient Heat Transfer (20)

Recently uploaded (20)

Simulation and Optimization of Cooling Tubes of Transformer for Efficient Heat Transfer