Content uploaded by Ali Shokor Golam
Author content
All content in this area was uploaded by Ali Shokor Golam on Jun 01, 2024
Content may be subject to copyright.
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences 117, Issue 2 (2024) 131-146
131
Journal of Advanced Research in Fluid
Mechanics and Thermal Sciences
Journal homepage:
https://semarakilmu.com.my/journals/index.php/fluid_mechanics_thermal_sciences/index
ISSN: 2289-7879
Evaluation of the Electrical Power Transformer Fins Design Technology:
Numerical Analysis and Experimental Validation
Ali Shokor Golam1,*
1
Department of Mechanical Engineering, College of Engineering, Mustansiriyah University, Baghdad 10047, Iraq
ARTICLE INFO
ABSTRACT
Article history:
Received 13 December 2023
Received in revised form 28 April 2024
Accepted 11 May 2024
Available online 30 May 2024
The study includes numerical analysis and experimental verification on an electrical
power distribution transformer (250 kVA, 11 kW, Oil Natural Air Natural). ANSYS Fluent
R3 2019 software was used to develop the numerical simulation model. The validity of
the numerical model was confirmed by comparing the results of the numerical model and
experimental data. The study aims to improve the efficiency and performance of electrical
power distribution transformers by proposing a design that reduces the temperature of
the transformer while maintaining its traditional size. Numerically, the effect of fin
geometry on the temperature and density of transformer oil was studied. Four different
fin designs were proposed and compared with the traditional design. According to the
results, all proposed designs contributed to improving the cooling performance of the
transformer compared to the traditional design. Design A is similar to the traditional
transformer design, with the only modification being manipulation of the fin length, and
achieves an average oil temperature reduction of 4 K. Design B showed the smallest
temperature drop of the four designs, with a 3 K drop. Designs C and D include ventilation
channels that match the shape of the fin, providing distinct design differences. The
difference between both designs relied on the fact that for design C, the orthogonally of
fin plates was retained. On the other hand, in design D, skewing of fin plates was
introduced. Design D proved to be the most effective in reducing the average oil
temperature, being reduced by 10 K. On the other hand, Design C reduced the average
oil temperature by 7 K.
Keywords:
Electrical power transformer; heat
transfer; numerical study; natural
convection; fin geometry
1. Introduction
The electrical transformer, a critical component of the electrical power transmission network,
plays a pivotal role in voltage manipulation for efficient electrical energy transmission. However,
complete consumption of the generated energy is not achievable due to inevitable losses. These
losses manifest in various forms, contributing to power dissipation within the electrical transformers,
resulting in elevated internal temperatures and subsequent performance degradation [1,2]. In many
countries, the efficiency of transformers decreases due to unsuitable operating environment and
overloading that exceeds the transformer's design capacity. Most electrical power distribution
* Corresponding author.
E-mail address: ali.shokor49@uomustansiriyah.edu.iq
https://doi.org/10.37934/arfmts.117.2.131146
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
132
transformers are designed to operate within a safe temperature between (55 - 65) °C [3]. According
to the Montsinger formula, the life span of the transformer is reduced by half when the temperature
increases by 8 °C above Its design temperature is due to damage to the insulators [4]. Recently,
studying the behaviour and performance of the cooling system of electrical power distribution
transformers has been a real challenge for many researchers. Hasan [5] studied the thermal
behaviour of an electrical power distribution transformer under the influence of air temperature.
nano oils were used to replace traditional transformer oil. Four types of nano-oils were prepared
using solid nanoparticles Cu, SiC, Al2O3 and TiO2. Of all the nano fluids examined, SiC oil nano fluid
provides the lowest transformer temperature. Rui [6] conducted a 3D simulation of an electrical
transformer to verify the use of different oils instead of conventional oil. The results indicate that
graphite, CNT, and diamond nano fluids have a higher heat transfer coefficient compared to the
original transformer oil. CNT-based nano fluids show superior heat transfer performance compared
to other nano fluids such as diamond and graphite. Hasan et al., [7] analyzed a proposed thermal
model for a transformer (250 kVA). Also, the effect of adding nanoparticles (CuO and Al2O3) on the
heat transfer process was examined. It turns out that the maximum temperature drop is 5%. When
using nano-oils. Pendyala et al., [8] investigated numerically the heat transfer and fluid motion
features of transformer oil and oil-based nano fluids. They found that there is better heat dissipation
in nano fluids compared to transformer oil. Also, it was shown that the overall heat transfer
performance in nano fluids decreases with any increase in density. Hannun et al., [9] investigated the
modelling of the power converter by using ANSYS-Fluent 17.1 code. A connection is made between
the power transformer and the air-to-ground heat exchanger to reduce the temperature of the
transformer. The results showed that the temperature of the transducer decreases with increasing
depth of the underground air heat exchanger and tube length. Medved et al., [10] investigated
improving the life of power transformers using ANSYS CFX through the use of insulation and cooling
oil. It turns out that natural and synthetic oils are more suitable for cooling. The thermal conductivity
of natural or synthetic oils is greater than that of mineral oils. Özgönenel et al., [11] conducted an
investigation using COMSOL to convert the distribution of an oil-insulated transformer to a gas-
insulated sulfur hexafluoride transformer. The new design provides safer transformation of security
risk sites such as mines, submarines and nuclear power plants. Arun and Manavendra [12] replaced
mineral oil with synthetic oil for cooling and insulation such as corn oil, palm oil, sunflower oil and
coconut oil. Among all the synthetic oils selected, the coolant fin outlet temperature reduction for
corn oil was found to be 2.3% lower than for conventional mineral oil. Si et al., [13] analyzed
numerically the leakage magnetic flux of metal components and oil tank of an electrical transformer
(400 kVA-15 kV/400 V). They found that the current is generated by the magnetic flux leaking inside
the coils (the core of the transformer), and as a result of the oil coming into contact with the metal
components, heat is transferred from the coils through the oil to the fins, so that the temperature of
the hot spot, which does not have a fixed location, decreases. Kanafiah et al., [14] Analyzed articles
that rely on the mathematical model to solve fluid flow problems using the thematic analysis program
ATLAS.ti 8. This review paper adopts a review of only 50 articles on the topic as final articles from
2015 to 2020. Studied will be useful in determining the expansion of the mathematical model for the
problems Related to fluid flow and convection as well as the geometric condition and appropriate
boundaries. Jackwin and Bharathi [15] modified the design in cooling tubes to give better turbulent
generation and improved heat transfer. A cooling tube with coaxial grooves and fins was found to be
the most effective. In addition, it was found that the natural heat flow of transmission oil can be
effectively simulated using fluid dynamics calculation due to the density variation with increasing
temperature. Ţălu and Ţălu [16] used the finite element analysis method to optimize the dimensions
of the front radiators of the cooling system of electric power transformers. The model was validated
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
133
by experimental results. As a result, optimization was obtained: reducing the cooling elements and
their size, reconfiguring the design, changing the dimensions of the oil tank, and optimal location of
the front and rear radiators on the oil tank. Chandak et al., [17] conducted a CFD analysis to
investigate the effect of radiation on transformer coolant heat dissipation. Close inspection provides
the possibility of increased heat dissipation even with the central coolant fins. Also, much higher heat
dissipation is provided in the final fins due to the entire face being opened to the periphery. These
results can be useful for designers to learn about the different effectiveness of fin configuration.
Rodriguez et al., [18] simulated experimental and dynamic computational measurements achieved
on a typical 30MVA power transformer. The main purposes of this study are to evaluate the cooling
capacity of current radiator design operating in ONAN mode. It has been shown that convective heat
transfer in plate is approximately 10 times slower than heat transfer in oil. Bachinger and Hamberger
[19] studied the operation of electrical power transformers under low ambient temperatures (-30 to
-50) degrees Celsius. It was noted that temperatures rose instead of decreased when using enhanced
cooling fans. Kaymaz [20] studied the thermal behaviour and flow pattern in a transformer radiator
filled with oils (Natural ester, metal and silicon). Natural ester oil was found to have the best heat
transfer and pressure drop performance. Abdolzadeh et al., [21] studied the thermal performance of
transformer work inside a closed and open space. The results showed that operating the transformer
inside a closed space causes a decrease in its efficiency due to lack of ventilation. Paramane et al.,
[22] examined the effect of fan mounting arrangement and air flow direction on the thermal
performance of power transformer radiators by using CFD analysis. The study examined four 3-
meter-high radiators, 30 0.52-meter-wide blades, and two 1-meter-diameter fans in the horizontal
and vertical airflow direction. It was found that fans placed on one side of the radiators compared to
the investigated arrangements lead to greater heat dissipation. Wu et al., [23] presented an accurate
numerical method to investigate the cooling effect of dry-type transformer unit with external heat
exchangers in comparison with laboratory experiments. It has been shown that CFD simulation tools
have a higher potential to be useful in practice either for developing design recommendations for
transformer cooling or to support the design and development of transformer products. Anishek et
al., [24] conducted numerical simulations of the coolant (oil and air) of natural power transformers
with the aim of determining the cooling capacity. Also, the optimal sections (spacing’s and lengths)
of the radiator are set. An optimal radiator design for measuring cooling capacity is presented and
simulated. The proposed design is 14% more efficient compared to the original design. Kumar and
Singhal [25] reviewed some studies on the efficiency improvements that can be made in pin fins to
enhance the heat transfer rate. It was concluded that there are many methods and techniques by
researchers in order to increase heat transfer by pin fins. Balaji et al., [26] emphasized the
commitment to take the original dimensions of the electrical transformer and take into account the
materials used for the body of transformers when studying their design. Conduct a thermal analysis
to compare the use of mild steel and aluminum for the transformer structure. Analysis proved that
mild steel has better strength than aluminum. They also found that mild steel has a good ability to
transfer heat through the fins surrounding the transformer body. Azbar et al., [27] conducted 3D
numerical modelling to test the effects of transformer shape and fins on the cooling performance of
ONAN type electrical power distribution transformer. The results showed that the best heat
dissipation performance and a significant 12% reduction in oil temperature compared to the
conventional transformer were obtained when using the transformer with a perforated trapezoidal
fin, hexagonal shape. Basher and Kadhem [28] analyzed fin shapes (trapezoidal, wavy, and triangular)
using Ansys Fluent 22 R1 software to examine their effect on the cooling efficiency of electrical
distribution transformers. They found that the highest effectiveness in reducing the surface and core
temperatures of the transformer was at the trapezoidal fin shape. Mahdi et al., [29] examined a
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
134
numerical model of an electrical transformer using ANSYS Fluent-15.0 software. Four different fin
designs were proposed. Design A with rectangular ventilation ducts has the same effect as the
traditional design. Slightly better performance was seen in the perforated rectangular design (Design
B). The best thermal performance was in Design C. Also, in terms of thermal distribution, the
conventional design had higher performance compared to Design D. Lee et al., [30] studied the design
technique of transformers with respect to pressure variation by simulating fluid-structure
interaction. The amount of oil expanding inside the corrugated fin was estimated for different load
cases, with a corrugated fin design proposed to control oil pressure variations inside. Under operating
conditions, the proposed transformer design reaches thermal and structural safety requirements.
Farhan et al., [31] recommended different fin shape under different flow powers to verify the optimal
design to enhance the flow and heat transfer features that can be used in transformer cooling. The
prevailing stresses decrease in a rectangular shape with a higher aspect ratio (h/s). Also, for cone-
shaped fins, they note a smooth temperature drop.
The literature review conducted above highlights that temperature rise plays a crucial role in the
efficiency of electrical transformers. Significant efforts have been made to enhance the thermal
performance, i.e., cooling rate, of transformers through various techniques. These techniques
include the use of heat exchangers, different types of oil, nanoparticle additives, and modifications
to the fins shape and transformer geometry. However, limited research has been conducted on the
potential of altering the design of the fins to increase the cooling rate of transformers. To address
this research gap, the primary objective of this study is to enhance heat transfer and optimize oil flow
patterns inside transformer fins by improving their design. This was achieved by developing a
numerical model of the thermal behaviour of a capacity 250 kVA, 11 kW type oil natural air natural
(ONAN) transformer using ANSYS Fluent, which was subsequently experimentally validated. Four
new fin designs were suggested, evaluated, and compared to the traditional transformer while
maintaining the same operational parameters. The ultimate goal is to prevent transformer
breakdown, extend their lifespan, and reduce the need for maintenance.
2. Problem Description
Efficient cooling of a transformer is crucial for maintaining temperature rise within acceptable
limits. During operation, the heat generated by active components is transferred to the oil, which in
turn conducts this heat to the walls of the transformer tank. The heat then dissipates from the walls
to the surrounding air through natural convection [32]. Fins are utilized to enhance the heat transfer
rate from the oil, flowing within the fins, to the external air. The current work includes selecting an
electrical power distribution transformer commonly used in electrical networks (250 KVA, Oil Natural
Air Natural (ONAN) type) as a case study. Figure 1 depicts an image of this transformer and a
schematic diagram. The components of the transformer include the transformer core, oil, and outer
shell. The core of the transformer includes a steel core carrying three oil-immersed copper coils, while
the transformer body is equipped with fins to increase the surface area for heat dissipation. The
dimensions of this electrical transformer are as follows: (0.7, 0.9 and 0.45) m represent height, length
and width transformer body, (0.6 and 0.2) m represent height and length of fins, the total number of
fins is 50, and the spacing between fins is 0.04m.
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
135
(a)
(b)
Fig. 1. ONAN type transformer (a) Traditional transformer image (b)
Schematic diagram
In this work, the thermal behaviour of four distinct transformer designs was examined using
numerical simulations and experimentally validated. These proposed transformer designs were
compared to a traditional design. To ensure a fair comparison, all case studies maintained nearly
identical oil volumes and external surface areas. Figure 2 illustrates the geometric characteristics of
the four different transformer designs (A, B, C, and D). While all designs share similar features, the
distinguishing factor lies in the fin shape. The total surface area of the fins in all proposed designs
amounts to 9.6 m2.
Design A
Design B
Design C
Design D
Fig. 2. The geometrical details of the four transformer designs
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
136
3. Numerical Model
3.1 Governing Equations
The oil flow within the transformer enclosure is simulated using a three-dimensional, steady-
state, and incompressible model. It is important to note that the properties of the oil are
temperature-dependent. The relationship between temperature and the oil properties can be
described using the following equations [33]
(1)
(2)
(3)
(4)
The model's governing equations can be summarized as follows [34]
Mass conservation
(5)
Momentum conservation
(6)
Energy conservation
(7)
Where ρ: Density, μ: Dynamic viscosity, CP: Specific heat, and k: Thermal conductivity, V: velocity,
P: pressure, β: thermal expansion coefficient, oil temperature K, : reference temperature K,
g: acceleration of gravity.
In order to simulate the heat source, the transformer core surface was treated as a constant heat
flux boundary. The heat source in the transformer represents the electrical loss from the transformer
core of 4215.3 W when the transformer electrical load is 300 kVA. As for the outer surface of the
transformer, it is assumed to be a thermal surface with a heat transfer coefficient of 19 W/ m2.K. This
heat transfer coefficient was determined in a previous study, which successfully validated the model
[34]. The air stream temperature was set to 308 K to represent the cooling effect of outside air on
the transformer. The transformer enclosure plate was considered to be made of a material with high
thermal conductivity, of negligible thickness.
3.2 Numerical Approach
To numerically solve the governing equations and associated boundary conditions, the
commercial software ANSYS FLUENT R3 2019 was used. The coupling of velocity and pressure is
solved using the SIMPLE algorithm, while a second-order upwind scheme is used for pressure. As for
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
137
turbulent dissipation and kinetic energy, a first-order solution was adopted. A relaxation technique
was applied to control the iteration process to ensure convergence. Relaxation drop factors were
determined as follows: 0.7 for momentum, 0.8 for energy equation, 0.3 for pressure and 1 for
viscosity. Two methods were used to verify the solution, the first was to track the average
temperature (less than 1°C every 3000 iterations), and the second was to keep the residual values
for the momentum and turbulent equations below 10-4, and below 10-5for the energy equations.
3.3 Grid Independence Test
In any numerical simulation, constructing a grid for the body is a crucial step where the
appropriate grid size is selected to ensure accurate results within the shortest possible time. Once
the geometry is constructed, the mesh is generated, and the boundary conditions are assigned,
multiple meshes need to be examined to determine an optimal grid system. Figure 3 illustrates the
outer view of the mesh employed in the computational model for design C. To assess mesh
independence, six different mesh sizes were evaluated, and the corresponding results for the average
temperature of the oil, considering an outer air temperature of 308 K, are presented in Table 1. The
analysis of Table 1 reveals that after the fourth mesh, further increases in the grid size do not
significantly affect the solution. Therefore, the fourth mesh is utilized for all numerical computations.
Fig. 3. Mesh generated for the computational model of the design C
Table 1
Grid size
Grid size
Average oil temperature (K)
983.456
347.34
1.263.571
342.41
1.518.523
340.35
1.824.261
338.02
2.091.629
337.31
2.301.814
337.72
4. Model Validation
To ensure the reliability of the generated model in simulating the thermal behaviour of the
proposed transformers, a comparison is made between the estimated steady-state temperatures at
a point situated 3 cm above and below the coil surface and the temperature readings obtained from
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
138
thermocouples placed at the same location. An in-situ test is conducted at the State Company for
Electrical Industries in Al-Waziriya, Iraq, to acquire temperature measurements while the
transformer is in operation. Figure 4 depicts the electrical transformer during the testing process.
The measured temperatures serve as validation data for the generated model. The test takes place
in a well-ventilated large building. Type K thermocouples positioned 3 cm above and below the core
of the transformer core are used to measure the oil temperature. A data logger is used to record the
thermal measurements directly to the computer every 3 hours. The thermocouple has a total
uncertainty of ±1 °C. The average air temperature surrounding the transformer, measured from four
different locations one meter from the transformer walls, is 299.5 K. The transformer is loaded at 1.2
times its rated load, which amounts to 300 kVA (since the capacity of the transformer is 250 kVA, the
test load is calculated as 250 kVA x 1.2 = 300 kVA). Testing continues until a stable temperature is
reached, which is indicated by a temperature rise of less than 1 K per 3 hours. Figure 5 displays the
transformer oil temperature measured during the test, showing a sharp increase initially until it
reaches the steady-state temperature. Figure 6 shows the average oil temperature over time for the
numerical and experimental results. The comparison showed a clear convergence between the
numerical results and experimental data, as the linear error rate did not exceed 9% for the average
temperature of the transformer oil. This convergence within an acceptable error rate determines the
reliability of the model.
Fig. 4. The electrical transformer during the testing
process
Fig. 5. Oil temperature in the upper and lower zone
during testing with constant air temperature 299.5
K, load 300 kVA
290
300
310
320
330
340
350
360
1357911 13
Temperature K
Time (houre)
Upper zone
Lower zone
Air temperture
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
139
Fig. 6. Comparison between numerical simulation
results and experimental data for the relationship
of average oil temperature with time
5. Results and Discussion
Numerical investigations are conducted to analyse the thermal behaviour and density variations
of the transformer's oil across four different transformer designs. Ratings are made taking into
account an ambient air temperature of 308 K and electrical losses from the transformer core, which
amount to 4215.3 W for a 300 kVA transformer load. Figure 7 illustrates the average oil temperature
for both the traditional transformer and the four proposed designs. The average oil temperature for
a traditional electrical transformer is recorded as 337 K. Design A, which shares similar design
parameters with the traditional transformer except for adjustments made to the fin length,
demonstrates an increase in the distance between fins. This modification prevents the convergence
of adjacent thermal layers formed on the surfaces of opposing fins. Although the decrease in average
oil temperature is marginal, reaching 333 K, the objective of studying Design A is not to deviate from
the conventional transformer design but to emphasize the crucial role played by fin length and
spacing in heat transfer. Design B exhibits an average oil temperature of 334 K, representing the
smallest temperature drop among the four designs and a reduction of 3 K compared to the standard
design. Despite the enlarged surface area of the fin in the upper hot region of the transformer, the
fin's shape impedes oil movement, which relies on the buoyancy phenomenon caused by density
differences, thereby reducing the heat transfer rate. Designs C and D present alternative designs
featuring ventilation channels that align with the fin's shape, aiming to ventilate the heated base of
the fins and keep them separate from direct contact with the transformer body. Design C conforms
to the shape of a conventional transformer fin with the addition of rectangular ventilation channels
that distance the base of the fin from the hot transformer body. The average oil temperature for
Design C was recorded as 331 K. I note that design D achieves an average temperature of 327 K. This
indicates that cooling is significant compared to the traditional design. This is attributed to two
factors: First, the fin design allowed for increased surface area for heat transfer in the upper hottest
region of the transformer, which is in line with the recommendations of the researcher Azbar et al.,
[27]. Second, large ventilation channels that conform to the shape of the fin helps keep the fin base
away from the hot transformer body. It also allows the circulation of oil within the fin to facilitate
heat exchange with the surrounding areas. The average oil temperature results for Designs C and D
indicate that the distance between the fin base and the transformer body allows direct heat transfer
290
300
310
320
330
340
350
360
1357911 13
Oil temperature K
Time (hour)
Numerical
Experimental
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
140
between the fin surface and the surrounding areas without the fin base being affected by the
temperature of the transformer body. Figure 8 indicates the variation of the average oil temperature
with time for all proposed designs in addition to the traditional transformer. It is noted that the
average oil temperature rises sharply at the beginning of the test until reaches stability over time,
and this temperature is considered the stable temperature of the electrical transformer.
Fig. 7. Average oil temperature for the
conventional transformer and the four proposed
designs
Fig. 8. Variation of average oil temperature with
time for conventional transformer and proposed
transformer designs
Figure 9 presents the variation in average oil density for both conventional transformers and the
proposed transformer designs. The behaviour of oil density exhibits a complete inverse relationship
to that of oil temperature, which aligns logically. The figure illustrates that Design D showcases the
lowest average oil density compared to the conventional transformer and other designs. This can be
attributed to the ability of Design D to facilitate oil flow into the fin cavities, allowing for heat
exchange away from the high-temperature region of the transformer, particularly in the upper area
known as the hot zone. With the presence of a large fin area in this region, a higher rate of heat
transfer is achieved, resulting in a decrease in oil temperature and, consequently, an increase in
density values. Figure 10 Shown the variation of average oil density with time for conventional
322
324
326
328
330
332
334
336
338
Traditional A B C D
Average oil temperature K
Design
310
315
320
325
330
335
340
1357911 13
Average oil temperature K
Time (hour)
Traditioal
Design A
Design B
Design C
Design D
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
141
transformer and proposed transformer designs. It is noted that the average oil density decreases
sharply at the beginning of the test until it reaches stability after seven hours, and this degree is
considered the stable oil density of the electrical transformer.
Fig. 9. Average oil density for conventional
transformer and proposed transformer designs
Fig. 10. Variation of average oil density with time
for conventional transformer and proposed
transformer designs
Figure 11 displays the transformer oil temperature limits throughout the entire computational
domain. The thermal behaviour of the designs studied showed a difference in temperatures
depending on the height of the transformer. Due to the natural convection current, heat is
transferred from the core of the transformer (coils) to the surroundings through the fins. During the
process of heat dissipation, it finds that the lowest oil temperature is at the bottom of the
transformer and increases gradually as we move toward the top. Therefore, the hottest area is at the
top and close to the surface of the heat source (the core of the transformer). This is due to the effect
of the buoyant force, which causes the low-density oil (hot oil) to rise to the upper part. This
phenomenon has been previously described by Tampinyo and Srikunwong [35], and Azbar and Jaffal
[36]. Regarding oil density, it is observed that it exhibits an inverse relationship with temperature.
Figure 12 displays the oil density contours for the traditional transformer and the four designs across
the entire computational domain. In general, the lower section exhibits higher oil density compared
to the upper section. The oil serves as an insulator for the coil while simultaneously transferring heat
845
850
855
860
865
870
Traditional A B C D
Average oil density 𝑘𝑔∕m3
Design
855
860
865
870
875
880
1 3 5 7 9 11 13
Average oil density 𝑘𝑔∕m3
Time (hour)
Traditional
Design A
Design B
Design C
Design D
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
142
to be dissipated into the surroundings. As the oil temperature rises, its density undergoes changes,
leading to a floating phenomenon.
Traditional electrical transformer
Design A
Design B
Design C
Design D
Fig. 11. The temperature contours of the transformer’s oil at the full computational domain
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
143
Traditional electrical transformer
Design A
Design B
Design C
Design D
Fig. 12. The oil density contours of the traditional transformer and the four designs at the full
computational domain
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
144
6. Conclusion
This paper presents a numerical study accompanied by experimental verification to investigate
the thermal behaviour of a three-dimensional model, specifically focusing on the impact of fin
geometry on the cooling efficiency of a capacity 250 KVA transformer type (ONAN). The current work
aims to improve the performance and efficiency of electrical transformers by proposing a design that
effectively reduces the temperature of the transformer while keeping its overall size similar to that
of a conventional transformer. The accuracy of the numerical model was evaluated by comparing the
numerical simulation model with the experimental data of the traditional transformer. The
simulation model is reliable, with a maximum error of about 9%. Four new designs are presented and
compared with the traditional transformer design. It can be concluded that the design and shapes of
the transformer fins play an important role in the heat transfer process, as their effect depends on
the large surface area of the hot components. Of the designs, Design D proved to be the most
effective in reducing the average oil temperature, achieving a temperature reduction of 10 K
compared to a traditional transformer. Design C, Design A, and Design B reduce the average oil
temperature by 7 K, 4 K, and 3 K, respectively.
Acknowledgement
This research was not funded by any grant. Also, author would like to thank the State Company for
Electrical Industries in Al-Waziriya, Ministry of Industries, Iraq and University of Al-Mustansiriyah
(www.uomustansiriyah.iq) Baghdad, Iraq for their continuous support and facilities.
References
[1] Agah, S. M., and Hossein Askarian Abyaneh. "Distribution transformer loss-of-life reduction by increasing
penetration of distributed generation." IEEE Transactions on Power Delivery 26, no. 2 (2011): 1128-1136.
https://doi.org/10.1109/TPWRD.2010.2094210
[2] Susa, Dejan, Matti Lehtonen, and Hasse Nordman. "Dynamic thermal modeling of distribution transformers." IEEE
Transactions on Power Delivery 20, no. 3 (2005): 1919-1929. https://doi.org/10.1109/TPWRD.2005.848675
[3] Perez, Joe. "Fundamental principles of transformer thermal loading and protection." In 2010 63rd Annual
Conference for Protective Relay Engineers, pp. 1-14. IEEE, 2010. https://doi.org/10.1109/CPRE.2010.5469518
[4] Hajidavalloo, Ebrahim, and Mohamad Mohamadianfard. "Effect of sun radiation on the thermal behavior of
distribution transformer." Applied Thermal Engineering 30, no. 10 (2010): 1133-1139.
https://doi.org/10.1016/j.applthermaleng.2010.01.028
[5] Hasan, Mushtaq Ismael. "Using the transformer oil-based nanofluid for cooling of power distribution transformer."
International Journal of Energy and Environment 8, no. 3 (2017): 229-238.
[6] Rui, Lim Lian. "Heat Transfer Performance of Oil-Based Nanofluids in Electric Transformers." Degree thesis,
Universiti Teknologi Petronas, 2015.
[7] Hasan, Ibtisam A., Sahar R. Fafraj, and Azhar K. Azeez. "Theoretical temperature distribution investigation in
electrical transformer by using nano-technology." Engineering and Technology Journal 34, no. 12A (2016): 2282-
2295. https://doi.org/10.30684/etj.34.12A.11
[8] Pendyala, Rajashekhar, Suhaib Umer Ilyas, Lian Rui Lim, and Narahari Marneni. "CFD analysis of heat transfer
performance of nanofluids in distributor transformer." Procedia Engineering 148 (2016): 1162-1169.
https://doi.org/10.1016/j.proeng.2016.06.619
[9] Hannun, Rafid M., Salman H. Hammadi, and Mohammed H. Khalaf. "Heat transfer enhancement from power
transformer immersed in oil by earth air heat exchanger." Thermal Science 23, no. 6 Part A (2019): 3591-3602.
https://doi.org/10.2298/TSCI171231116H
[10] Medved, D., M. Beluscak, and J. Zbojovsky. "Solution of thermal field around transformer." In Eighth International
Scientific Symposium on Electrical Power Engineering, Elektroenergetika, pp. 140-143. 2015.
[11] Özgönenel, Okan, David Thomas, and Ünal Kurt. "SF6 gas-insulated 50-kVA distribution transformer design."
Turkish Journal of Electrical Engineering and Computer Sciences 26, no. 4 (2018): 2140-2150.
https://doi.org/10.3906/elk-1708-28
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
145
[12] Arun, N. R., and G. Manavendra. "Performens and Analysis of Transformer Radiator with Different Types of Oils
Using CFD." International Journal of Innovative Research in Science, Engineering and Technology 5, no. 12 (2016):
20835-20843.
[13] Si, Wen-Rong, Chen-Zhao Fu, Xu-Tao Wu, Xiu Zhou, Xiu-Guang Li, Yi-Ting Yu, Xiao-Yu Jia, Jian Yang, and Hervé
Laurent. "Numerical study of electromagnetic loss and heat transfer in an oil-immersed transformer."
Mathematical Problems in Engineering 2020 (2020): 1-13. https://doi.org/10.1155/2020/6514650
[14] Kanafiah, Siti Farah Haryatie Mohd, Abdul Rahman Mohd Kasim, Syazwani Zokri, and Nur Syamilah Arifin. "A
Thematic Review on Mathematical Model for Convective Boundary Layer Flow." Journal of Advanced Research in
Fluid Mechanics and Thermal Sciences 86, no. 2 (2021): 107-125. https://doi.org/10.37934/arfmts.86.2.107125
[15] Jackwin, Vincent K., and V. V. Prathibha Bharathi. "Simulation and Optimization of Cooling Tubes of Transformer
for Efficient Heat Transfer." International Journal of Advanced Engineering, Management and Science 3, no. 2
(2017): 141-146. https://doi.org/10.24001/ijaems.3.2.23
[16] Ţălu, Ştefan D. L., and Mihai D. L. Ţălu. "Dimensional optimization of frontal radiators of cooling system for power
transformer 630 kVA 20/0.4 kV in terms of maximum heat transfer." UPB Scientific Bulletin, Series C 72, no. 4 (2010):
249-260.
[17] Chandak, V., S. B. Paramane, W. V. d Veken, and Joris Coddé. "Numerical investigation to study effect of radiation
on thermal performance of radiator for ONAN cooling configuration of transformer." In IOP Conference Series:
Materials Science and Engineering, vol. 88, no. 1, p. 012033. IOP Publishing, 2015. https://doi.org/10.1088/1757-
899X/88/1/012033
[18] Rodriguez, Gustavo Ríos, Luciano Garelli, Mario Storti, Daniel Granata, Mauro Amadei, and Marcelo Rossetti.
"Numerical and experimental thermo-fluid dynamic analysis of a power transformer working in ONAN mode."
Applied Thermal Engineering 112 (2017): 1271-1280. https://doi.org/10.1016/j.applthermaleng.2016.08.171
[19] Bachinger, F., and P. Hamberger. "Thermal measurement of an ester-filled power transformer at ultra-low
temperatures: steady state." Procedia Engineering 202 (2017): 130-137.
https://doi.org/10.1016/j.proeng.2017.09.700
[20] Kaymaz, Özben. "Investigation of oil flow and heat transfer in transformer radiator." PhD diss., Izmir Institute of
Technology (Turkey), 2015.
[21] Abdolzadeh, M., A. Khoshabi, and M. A. Mehrabian. "Thermal analysis of an electric power transformer inside an
enclosure-a case study." International Journal of Ambient Energy 38, no. 3 (2017): 250-258.
https://doi.org/10.1080/01430750.2015.1086678
[22] Paramane, Sachin B., Wim Van der Veken, Atul Sharma, and Joris Coddé. "Effect of fan arrangement and air flow
direction on thermal performance of radiators in a power transformer." Journal of Power Technologies 97, no. 2
(2017): 127.
[23] Wu, Wei, Yong Wang, and Zepu Wang. "CFD simulation on heat exchanger cooled dry-type transformers." In
Proceedings of the 2nd World Congress on Electrical Engineering and Computer System and Science (EECSS'16), vol.
10. 2016. https://doi.org/10.11159/eee16.143
[24] Anishek, S., R. Sony, J. Jayadeep Kumar, and Pradeep M. Kamath. "Performance analysis and optimisation of an oil
natural air natural power transformer radiator." Procedia Technology 24 (2016): 428-435.
https://doi.org/10.1016/j.protcy.2016.05.059
[25] Kumar, Naveen, and Udit Singhal. "A Review Paper on Fin Efficiency Enhancement." International Journal of Trend
in Scientific Research and Development 4, no. 1 (2019): 901-904.
[26] Balaji, Dwarapudi Sai, Dwarapudi Ashok, Gondesi Gopala Reddy, Gandi Upendra, and Nakka Suneel Kumar. "Design
and Thermal Analysis on Transformer Fin Using CFD." International Journal of Research in Engineering, Science and
Management 5, no. 6 (2022): 55-57.
[27] Azbar, Nawras Mohammed, Hayder Mohammad Jaffal, and Basim Freegah. "Enhancement of the thermal
performance characteristics of an electrical power transformer." Engineering Science & Technology (2021): 1-21.
https://doi.org/10.37256/est.212021487
[28] Basher, Hadi O., and Fatima M. Kadhem. "Numerical assessment of the impact of fin configuration on the cooling
performance of a power transformer." Wasit Journal of Engineering Sciences 11, no. 2 (2023).
https://doi.org/10.31185/ejuow.Vol11.Iss2.445
[29] Mahdi, Mustafa S., Anees A. Khadom, Hameed B. Mahood, Mahmood Abdul Razak Yaqup, Jammal M. Hussain,
Khalid I. Salih, and Hussein A. Kazem. "Effect of fin geometry on natural convection heat transfer in electrical
distribution transformer: numerical study and experimental validation." Thermal Science and Engineering Progress
14 (2019): 100414. https://doi.org/10.1016/j.tsep.2019.100414
[30] Lee, S., J. Y. Lee, J. C. Yun, J. Y. Park, and J. H. Woo. "Development of thermal and structural design technology for
a hermetically sealed oil transformer." Structures Under Shock and Impact XII 126 (2013): 179.
https://doi.org/10.2495/SU120161
Journal of Advanced Research in Fluid Mechanics and Thermal Sciences
Volume 117, Issue 2 (2024) 131-146
146
[31] Farhan, Muhammad, Muhammad Saad Hameed, Hafiz Muhammad Suleman, and Muhammad Anwar. "Heat
transfer enhancement in transformers by optimizing fin designs and using nanofluids." Arabian Journal for Science
and Engineering 44, no. 6 (2019): 5733-5742. https://doi.org/10.1007/s13369-019-03726-9
[32] IEEE Standards Association. IEEE Standard Test Procedure for Thermal Evaluation of Insulation Systems for Liquid-
Immersed Distribution and Power Transformers - Redline. IEEE Std C57.100-2011 (Revision of IEEE Std C57.100-
1999), pp. 1-55. IEEE, 2012.
[33] Gong, Ruohan, Jiangjun Ruan, Jingzhou Chen, Yu Quan, Jian Wang, and Cihan Duan. "Analysis and experiment of
hot-spot temperature rise of 110 kV three-phase three-limb transformer." Energies 10, no. 8 (2017): 1079.
https://doi.org/10.3390/en10081079
[34] Phu, Nguyen Minh, and Pham Ba Thao. "Thermohydraulic performance of a fin and inclined flat tube heat
exchanger: A numerical analysis." CFD Letters 13, no. 7 (2021): 1-12. https://doi.org/10.37934/cfdl.13.7.112
[35] Tampinyo, Panitan, and Chainarong Srikunwong. "Numerical Simulation of Heat Transfer Process in Oil-Immersed
Transformer." In The 5th TSME International Conference on Mechanical Engineering. 2014.
[36] Azbar, Nawras Mohammed, and Hayder Mohammad Jaffal. "Experimental Study of the Thermal Performance
Behaviour of Electric Power Transformers." Journal of Engineering and Sustainable Development, First Online
Scientific Conference for Graduate Engineering Students (2020). https://doi.org/10.31272/jeasd.conf.1.51
