JOURNAL OF OPTOELECTRONICS AND ADVANCED MATERIALS Vol. 10, No. 5, May 2008, p.

1149 - 1158

The state of the art in engineering methods for

transformer design and optimization: a survey

ELEFTHERIOS I. AMOIRALIS*, MARINA A. TSILIa, PAVLOS S. GEORGILAKIS

Department of Production Engineering & Management, Technical University of Crete, GR-73100, Chania, Greece
a
Faculty of Electrical & Computer Engineering, National Technical University of Athens, GR-15780, Athens, Greece

The complexity of transformer design demands reliable and rigorous solution methods. Survey of current research reveals
the continued interest in application of advanced techniques for transformer design optimization. This paper summarizes
some of the most important developments in this research area. The main purpose is to provide a synthesis of the
published research in this field and stimulate further research interests and efforts in the respective topics.

(Received March 13, 2008; accepted May 5, 2008)

Keywords: Survey, Transformers, Transformer Design, Transformer Design Optimization, Numerical Techniques,
Artificial Intelligence, Equivalent Circuit Models, Experimental Models

1. Introduction for introducing mutual coupling between electric circuits.

Transformers are extensively used in electric power
Transformer design is a complex task in which systems to transfer power by electromagnetic induction
engineers have to ensure that compatibility with the between circuits at the same frequency, usually with
imposed specifications is met, while keeping changed values of voltage and current. Transformers are
manufacturing costs low. Moreover, the design one of the primary components for the transmission and
methodology may vary significantly according to the distribution of electrical energy. Their design results
transformer type (distribution, power or instrument mainly from the range of application, the construction, the
transformer) and its operating frequency (ranging between rated power and the voltage level.
50/60 Hz and a few MHz), while many alterations
according to the core constructional characteristics, the
cooling method, or the type of the magnetic material may 2.1 Transformer types
be encountered [1.2]. This paper provides an overview of
research, development and application of various Different kinds of transformer types may be
computational methods for transformer design, based on encountered, according to their use, their cooling method
an extensive number of published papers. The review is or the construction of their magnetic circuit. A major
divided into two major sections: research efforts focusing classification is realized according to the power and
on the prediction and/or optimization of specific voltage ratings: transformers with a rated power up to 2.5
transformer characteristics (mainly losses and short-circuit MVA and a voltage up to 36 kV are referred to as
impedance) and techniques adopted for global transformer distribution transformers, while all transformers of higher
design optimization, taking into account all the relevant ratings are classified as power transformers. Power
performance parameters. transformers may be further classified according to their
The paper is organized as follows: Section 2 describes scope of application, according to the followings.
the various transformer types that are present in the Transformers that are directly connected to the generator
relevant bibliography as well as the main considerations of a power station are called generator transformers. Their
during the transformer design process. Section 3 includes power range goes up to far above 1000 MVA and their
the survey overview of research dedicated to transformer voltage range extends to approximately 1500 kV. The
characteristics, while Section 4 provides an overview of connection between the different high-voltage system
the research conducted on transformer design levels is made via network transformers (network
optimization. Finally, Section 5 concludes the paper. interconnecting transformers), which are mainly
autotransformers, i.e., transformers where the primary and
secondary winding of each phase have a common section.
2. Transformer design Their power range exceeds 1000 MVA and their voltage
range exceeds 1500 kV. Distribution transformers are used
A transformer has been defined by ANSI/IEEE [3] as in the distribution networks in order to transmit energy
a static electric device consisting of a winding, or two or from the medium voltage network to the low voltage
more coupled windings, with or without a magnetic core, network of the consumers. In addition, there are various
1150 Eleftherios I. Amoiralis, Marina A. Tsili, Pavlos S. Georgilakis

special purpose transformers such as converter structure of the survey overview presented in the next
transformers, test transformers, instrument transformers or Sections.
telecommunications transformers, which can be both in the
range of power transformers and in the range of
distribution transformers as far as rated power and rated
voltage are concerned.
As far as the cooling method is concerned,
transformers may be designed either as oil-immersed or
dry type transformers. In oil-immersed transformers, the
insulating medium is oil while in dry type transformers,
the cooling is implemented with natural air circulation.
The identification of oil-immersed transformers according
to the cooling method is expressed by a four-letter code.
The first letter expresses the internal cooling medium in
contact with the windings. The second letter identifies the
circulation mechanism for internal cooling medium. The
third letter expresses the external cooling medium. The
fourth letter identifies the circulation mechanism for
external cooling medium. For example, if the internal
cooling medium is mineral oil, which is circulated with
natural flow, and the external cooling medium is air,
which is circulated with natural convection, then this
cooling method is coded as ONAN (Oil Natural Air
Natural). In power transformers, various cooling methods
are used including oil circulation by pumps, or forced air
circulation by fans, or both of the above. As a result, four
cooling methods exist: ONAF (Oil Natural Air Forced),
OFAN (Oil Forced Air Natural) (OFAF Oil Forced Air
Forced) and OFWF (Oil Forced Water Forced).
Combinations like ONAN/ONAF, ONAN/OFAN or Fig. 1. Transformer survey structure.
ONAN/OFAF are also applicable [4].
Transformer magnetic circuit is constructed in either a
shell or a core structure. They are distinguished from each
other by the manner in which the primary and secondary 3. Research dedicated to specific transformer
coils are placed around the laminated steel core. The shell characteristics
type transformer is one where the windings are completely
surrounded by transformer steel in the plane of the coil. In The numerous computational methods and
core type, the windings surround the laminated iron core. engineering models proposed for transformer analysis and
There are two different technologies for stacking the the accurate prediction of their characteristics can be
sheets of the magnetic material of the core, providing a roughly categorized into four main groups:
further distinction to a) stack core transformers, where the 1. Numerical techniques that consist some of the most
layers of the sheets of the magnetic material are placed one widely used tools for transformer simulation. Among
over the other and the vertical and the horizontal layers are the proposed techniques of this group, the Finite
over lapped, and b) wound core transformers, where the Element Method (FEM) is the most prevalent one.
magnetic circuit is of shell type and the sheets are wound. 2. Stochastic methods including Artificial Intelligence
Multi-winding transformers, as well as poly-phase (AI) techniques, such as Genetic Algorithms (GAs),
transformers, can be made in either shell or core type which have seen increased usage in the transformer
designs, with a magnetic circuit that consists of five (in design area over the last few years.
shell-type transformers) or three legs (in core-type 3. Improved versions of the transformer equivalent
transformers), respectively. circuit, in order to include semi-empirical
descriptions of the core and winding characteristics
that affect the accuracy of calculations. Its use is still
2.2 Transformer design considerations common in the manufacturing industry, due to its
simplicity and its ability to provide reliable results,
Transformer design must take into account numerous especially in cases of standardized geometries.
performance parameters and technical constraints. The 4. Experimental methods, combining data provided by
research in the relevant literature may deal with each one measurements with analytical or other methods, in
of these parameters separately, or concern the overall order to provide efficient models for the accurate
transformer optimization. Fig. 1 presents the main representation of certain transformer characteristics.
categories of the literature survey, which define the
 The state of the art in engineering methods for transformer design and optimization: a survey 1151

3.1 No-load losses addition, Adly [34] presented an analytical investigation of

the various transformer losses resulting from semi-rotating
No-load losses are the continuous losses of a flux excitation. The main feature of this analysis is that
transformer, regardless of load, namely they exist core magnetic properties are accurately represented and
whenever the unit is energized [5]. No-load losses are also simulated using recently developed vector Preisach-type
called iron or core losses because they are mainly a models of hysteresis.
function of the core materials. The two main components Experimental methods, combining data provided by
of no-load losses are eddy currents and hysteresis. measurements with analytical or other methods, in order to
Hysteresis describes the memory of a magnetic material. provide efficient models for the accurate representation of
More force is necessary to demagnetize magnetic material hysteresis and power losses in the laminations of power
than it takes to magnetize it; the magnetic domains in the transformers are proposed in [35,36], based on data
material resist realignment. Eddy current losses are small supplied from steel manufacturer. Mosses [37] reported
circulating currents in the core material. The steel core is a results in which localised flux density and losses have
conductor that carries an alternating magnetic field, which been measured experimentally in model cores and
induces circulating currents in the core. These currents compared with data obtained from a FEM analysis of the
through the resistive conductor generate heat and losses. same core geometry. In addition, Girgis et al. [38] carried
Cores are typically made from cold-rolled, grain-oriented out an analytical study in an attempt to determine the
silicon steel laminations. magnitude of the effects of a number of core production
The FEM has been extensively employed in the no- attributes. Albach et al. [39] present a practical method for
load losses prediction problem. The application of two- predicting the core losses in magnetic components for an
dimensional (2D) FEM in the calculation of transformer arbitrary shape of the magnetizing current. Furthermore,
core losses is realized in [6,7], where tools for the Dolinar [40] determined a magnetically nonlinear iron
calculation of stray and eddy losses, and hysterisis are core model of a three phase three-limb transformer and
developed. Detailed modeling of the core magnetic compared it with the classical saturated iron core model
properties based on measurements are presented in [8-12]. [41].
In order to achieve more detailed analysis and computation Many researchers have studied rotational iron losses
of the transformer magnetic field, the three-dimensional over the years and a number of techniques for making
(3D) analysis becomes necessary, as proposed in [13]. measurements have been described in the literature.
Moreover, Koppikar et al. 0 describe details of statistical Stranges and Findlay [42] described an apparatus capable
analysis used in conjunction with 2-D FEM, for of determining iron losses due to rotational flux. In
quantifying the effect of various factors affecting flitch addition, Findlay et al. [43] and Davies and Moses [44]
plate loss along with results of 3-D FEM simulations have carried out experiments on various samples in order
carried on slotted and laminated flitch plates. Furthermore, to test the hypothesis that different stacking patterns of
a rigorous analytical study using the finite difference grain oriented silicon steel laminations, cut at small angles
method for magnetic field calculation is performed by to the rolling direction, can reduce the iron core losses in
several authors in the literature in order to understand the power transformers. In addition, Marketos and Meydan
role of joints in determining the performance of cores of [45] introduced a novel method of fabricating consolidated
power and distribution transformers [15] and to accurately stacks of electrical steel into single-phase transformer
calculate the 3D spatial distribution, components, and total cores, which can considerably speed up the time required
core losses in power transformer stacked cores [16,17]. to build the cores by attempting to further reduce the
In addition, various studies [18-25] explore the local amount of flux that deviates from the rolling direction at
flux distribution in transformer cores as a function of joint the corners of transformer cores.
design and its relevance for power loss and noise. It is
worth noting that although transformer joint air gaps have
been well studied using FEM, they are seldom taken into 3.2 Load losses
account in circuit models [26].
The application of AI in loss evaluation is addressed Load losses result from load currents flowing through
in [27,28], where the no-load losses as a function of core the transformer [5]. Load losses are also called copper or
design parameters are predicted by means of Artificial wire or winding losses. The two components of the load
Neural Networks (ANNs). Georgilakis et al. [29,30] also losses are the I2R losses and the stray losses. I2R losses are
used ANNs to reduce the iron losses of assembled based on the measured DC resistance, the bulk of which is
transformers while optimizing the production process of due to the winding conductors, and the current at a given
individual cores using Taguchi methods. In addition, a load. The stray losses are a term given to the accumulation
combination of three AI methods is presented in [31], of the additional losses experienced by the transformer,
namely ANNs, Decision Trees (DTs) and GAs, in order to which includes winding eddy losses and losses due to the
minimize iron losses during manufacturing of wound-core effects of leakage flux entering internal metallic structures.
distribution transformers. Auxiliary losses refer to the power required to run
Accurate calculation of losses is also provided by the auxiliary cooling equipment, such as fans and pumps, and
development of improved equivalent circuits, as proposed are not typically included in the total losses.
in [32], which is able to predict losses under any A review of about 50 papers were conducted by
circumstances. Moreover, a final equivalent circuit of Kulkarni and Khaparde [46], which have dealt with one or
sufficient accuracy, combining both hysteresis and eddy more components of stray loss from the point of view of
current losses contributions is proposed in [33]. In estimation and reduction. In this case, stray losses include
1152 Eleftherios I. Amoiralis, Marina A. Tsili, Pavlos S. Georgilakis

eddy and circulating current loss in windings, losses in order to simplify the time and complexity of the
flitch plate, core edge loss, loss due to high current field, calculations required in automated design process.
and frame and tank losses. Stochastic methods are also employed for solving
Moreover, a short methodological survey is carried problems of this category, as by Thilagar and Rao [67],
out by Krawczyk and Turowski [47], showing the need of who suggested an exact equivalent circuit model for the
eddy current analysis in electric devices. estimation of all impedance parameters of three winding
The FEM analysis is quite commonly used for the transformers, with the use of GA. The suggested method
eddy loss calculations [48,49]. Pern and Yeh [50] are also estimates geometrically a complex parameter, that is,
engaged in the formulation of an A-V finite element mutual leakage between secondary and tertiary windings.
method to simulate the electromagnetic field and current
distribution in the windings of power transformers with
non-negligible circulating current. In addition, the foil- 3.4 Inrush current
winding eddy loss is estimated by Ram [51]. Furthermore,
the eddy current field due to both windings and heavy Transformer inrush currents are high-magnitude,
current leads in large power transformers is analyzed in harmonic-rich currents generated when transformer cores
[52] by using the improved T-Ω method. are driven into saturation during energization. These
When a three-phase three-limb core type transformer currents have undesirable effects, including potential
is subjected to the load-loss measurement test, it is found damage or loss-of-life to the transformer, protective relay
that the losses and currents of the phases are asymmetrical misoperation, and reduced power quality on the system
(and unequal). Reference [53] attempts to give a [68]. Inrush current prediction is therefore another
comprehensive explanation for the asymmetry in the loss important issue during transformer design and various
values of the three phases during the load-loss test. A approaches to deal with it are present in the technical
three-phase transformer has been analyzed using a literature.
comprehensive 3-D time-harmonic FEM analysis. Numerical techniques are present in the above context
of inrush current prediction, as in [69], where 2D FEM is
applied to three-legged power transformers for the
3.3 Leakage field and short-circuit impedance evaluation of forces on the windings due to inrush current
and their comparison to the respective short-circuit forces.
The calculation of transformer leakage flux is a However, the majority of the methods used for inrush
prerequisite to the calculation of reactance, short-circuit current simulations are based on the derivation of
impedance, short-circuit forces and eddy current losses. appropriate equivalent circuits, taking into account the
The finite element method has been extensively core geometry [70] and structural characteristics [71], the
employed in transformer leakage field evaluation. The first core material nonlinear characteristics [72] or using real-
research attempts, presented over three decades ago, [54], time measurements [73]. Artificial neural networks have
focused on 2D modeling, due to the restricted performance also been employed for the computation of inrush current
abilities provided by the early development of personal and forces [74].
computers. 2D FEM remains an efficient tool for leakage
field and short-circuit impedance evaluation and more
recent developments in the prediction of these parameters 3.5 Stresses and dynamic behavior under short
are presented in the literature [55]. Although the 2D circuits
modeling is convenient and useful in some design
problems, it can be found insufficient for detailed analysis The short-circuit current in a transformer creates
and computation of the transformer magnetic field, and enormous forces on the turns of the windings. The short-
therefore the three-dimensional (3D) solution becomes circuit currents in a large transformer are typically 8 to 10
necessary [56-60]. The boundary-element method is times larger than rated and in a small transformer are 20 to
another numerical technique extensively used for 25 times larger than rated. The forces on the windings due
electromagnetic problems [61]. The main attraction of this to the short-circuit current vary as the square of the
method is the simplicity of the data required to solve these current, so whereas the forces at rated current may be only
problems, along with the high accuracy obtained with a few newtons, under short-circuit conditions these forces
boundary elements. Moreover, the combination of can be tens of thousands of newtons. These mechanical
boundary and finite elements is another widely used and thermal stresses on the windings must be taken into
numerical field analysis technique presenting significant consideration during the design of the transformer.
advantages in transformer leakage field modeling [62,63]. Transformer behavior under short-circuits is one of the
To overcome the main numerical techniques major concerns during their design, since the ability to
drawback, namely the complexity of the required mesh overcome the resulting stresses and currents for external or
size, especially in 3D configurations, alternative leakage internal faults of a certain time duration without significant
field evaluation models have been proposed, with the use consequences on their operation is a requirement often
of a 3D reluctance network method [64], falling into the present in the international technical standards. The
category of equivalent circuit representation. An accurate representation of this behavior is mainly realized
alternative method of transformer leakage field calculation through numerical techniques due to their prevalence in
is based on simplified analytical formulas [65,66], a the transformer detailed magnetic field calculation, [75-
method often employed by transformer manufacturers in 80]. Equivalent circuit approaches are also employed,
although their application is usually coupled to some
 The state of the art in engineering methods for transformer design and optimization: a survey 1153

detailed calculation, often based to numerical method [81], temperature rise. The temperature rise in a transformer is
in order to ensure better representation of the phenomena intrinsic to that transformer at a fixed load. The design of
occurring during short-circuits. the cooling system is based on the hot-spot temperature
value, and different methods for its prediction are
3.6 Transformer noise proposed in the literature, along with the overall
temperature distribution prediction, according to the
Transformers located near a residential area should transformer cooling method.
have sound level as low as possible. The design and the The finite difference method is proposed by Pierce
manufacture of a transformer with low sound level require [94] for hot-spot temperature prediction in dry-type
in-depth analysis of noise sources. Core, windings and transformers. 2D FEM thermal calculation is proposed in
cooling equipment are three main factors of noise, with the [95] for the calculation of core hot-spot temperature in
first factor the paramount one. To determine a method for power and distribution transformers. Most recent trends in
the optimum design of the noise-reduction transformer, thermal modeling employ coupled electromagnetic-
noise attenuation of a simple structured prototype thermal finite element models [96,97].
transformer that utilizes C-cores is quantitatively discussed Appropriate equivalent circuits are developed in [98]
based on the equivalent circuit analysis [82,83]. Similar for the core rise temperature calculation and power
work is presented in [84]. Finally, a recently developed transformer thermal distribution [99-101]. In [102] and
calculation scheme for the computer modelling of the [103] the parameters of the proposed oil-immersed
load-controlled noise of oil-insulated three-phase power transformer thermal model derive with the use of genetic
transformers is presented in [85]. This modelling scheme algorithms. Dynamic thermal models are also proposed in
allows the precise and efficient computation of the coupled the literature [104-106], taking proper account of oil
electromagnetic, mechanical and acoustic fields. The viscosity changes and loss variation with temperature.
equations are solved using the FEM as well as the Particle swarm method and neural networks are also
boundary element method (BEM). encountered in the relevant research field [107,108].
Experimental investigation of thermal distribution of
3.7 Transformer insulation cast-resin and oil-filled transformers is carried out in [109]
and [110], respectively. Experimental data are used for the
The insulation of a transformer is linked to its ability improvement of analytical equations that predict thermal
to withstand surge phenomena and overvoltages likely to distribution of liquid filled transformers in [111].
occur during its operation. For this purpose, the related Last but not least, in order to improve power
work may deal with the analysis of such phenomena, so as transformer reliability, a special focus has been carried out
to design an adequate transformer insulation system. Other on insulating materials and especially on insulating oils.
factors that affect transformer insulation life are vibration The most commonly used liquid in power transformers is
or mechanical stress, repetitive expansion and contraction, mineral oil due to its low price and its good properties
exposure to moisture and other contaminants, and [112]. However, the performance of mineral oil starts to be
electrical and mechanical stress due to over-voltage and limited. Numerous activities have been initiated to try to
short-circuit currents. improve the properties of mineral oil or to find other
Numerical methods are more scarcely applied for the substitute liquids. Natural esters or vegetable oils have
simulation of the above phenomena, and are mainly used been used successfully as transformer dielectric coolants,
for the calculation of the transformer electric field [86]. On alternatively to conventional mineral oils [113-117]. Their
the other hand, the majority of research is based on application offers some advantages, such as safety against
equivalent circuit representation for transformer analysis a fire incident, environmental friendliness and improved
under overvoltages, respective to their geometrical transformer performance. Moreover, mixtures consisting
characteristics [87], their dynamic behavior [88], their of mineral oil and two other kinds of insulating liquids
frequency response characteristics [89] or the (namely silicon and synthetic ester oils) are investigated in
characteristics of the network where they are connected [118,119]. Finally, the “retrofilling” application, in this
[90]. Other attempts to model the insulation structure of case replacing mineral oil with natural esters, is also
transformers and the quantitative analysis of its dielectric presented in [120], as a method to extend the remaining
response are also encountered [91,92]. Finally, the thermal life of an aging transformer.
exploitation of measurements and the experience from the
impact of the operational characteristics on the insulation
life assessment is often one of the main methods to design 3.9 Transformer DC bias
an insulation system and predict its ability to withstand
any transient phenomena during the transformer life [93]. Direct Current (DC) can flow in Alternating Current
(AC) power lines if a DC potential difference exists
3.8 Transformer cooling between the various grounding points. Such a difference
can be caused by a geomagnetic storm or the injection of
Transformer cooling is one of the most important DC current by one of the ground electrodes of a DC link
parameters governing a transformer’s life expectancy. The [121]. Direct current flowing through the earthed neutrals
total temperature is the sum of the ambient and the of transformer windings causes a DC component in the
1154 Eleftherios I. Amoiralis, Marina A. Tsili, Pavlos S. Georgilakis

magnetising current. Owing to non-linearity, the waveform superconducting coil that simulated the inner secondary
of this current is strongly distorted. The prediction and winding of a high-superconducting traction transformer is
impact of this phenomenon has been studied with finite presented in [147,148]. Moreover, the magnetization
element method [122-124] and equivalent magnetic losses in HTS pancake windings according to the
circuits [125,126]. operating temperature, is discussed in [149].
There is a considerable progress in the technology of
gas immersed transformers in the last decade. Unlike the
3.10 Transformer monitoring and diagnostics oil-immersed transformers, they have SF6 gas for the
insulation and cooling purposes [150,151].
Despite the fact that monitoring and diagnostics are
not part of the transformer design process, they are
relevant to the main design considerations. For this 4. Transformer design optimization
purpose, an overview of some key works dealing with the
characteristics presented in the previous Sections is The difficulty in achieving the optimum balance
provided in the present Section. between the transformer cost and performance is a
AI techniques and stochastic methods are prevailing complicated task, and the techniques that are employed for
in the present category. Neural networks are encountered its solution must be able to deal with the design
in the majority of the research dealing with transformer considerations of Section 3, so as to provide a design
fault diagnosis [127-131]. They are also employed for oil- optimum, while remaining cost-effective and flexible. The
immersed distribution transformer monitoring in [132]. research associated with design optimization is therefore
Furthermore, Stochastic Petri Nets are used for the more restricted involving different mathematical
simulation of the fault diagnosis process of oil-immersed optimization methods.
transformers and the definition of the actions followed to Techniques that include mathematical models
repair the transformer [132]. Particle swarm method is employing analytical formulas, based on design constants
used for winding deformation identification in [133]. and approximations for the calculation of the transformer
Finally, fuzzy systems and expert systems are proposed in parameters are often the base of the design process
[135] and [136] for the gas analysis and insulation adopted by transformer manufacturers [152]. Artificial
monitoring, respectively. Intelligence techniques have been extensively used in
order to cope with the complex problem of transformer
design optimization, such as genetic algorithms (GAs) that
3.11 Recent trends in transformer technology have been used for transformer cost minimization [153],
performance optimization of cast-resin distribution
In the last decade, rapid changes and developments transformers with stack core technology [154] or toroidal
have been made in the field of the transformer design. core transformers [155]. Neural network techniques are
Continuous efforts are directed at developing also employed as a means of design optimization as in
improved electrical steels with lower iron losses for [156] and [157], where they are used for winding material
energy-efficient transformers. It is well known that low selection and prediction of transformer losses and
magnetic losses of amorphous material are attributable to reactance, respectively. Deterministic methods may also
the material's amorphous condition and small thickness of provide robust solutions to the transformer design
the ribbon [137,138]. The core losses can be limited by optimization problem. In this context, the deterministic
insulating coatings [139], and various types of coatings method of geometric programming has been proposed in
have been developed for application to both fully [158] in order to deal with the design optimization
processed and semiprocessed electrical steels. In addition, problem of both low frequency and high frequency
Matsuura et al. [140] presented long-term property transformer. The overall manufacturing cost minimization
reliability for iron-based amorphous alloy for use in oil- is scarcely addressed in the technical literature, and the
immersed transformer cores. main approaches deal with the cost minimization of
Advent of high-temperature superconducting (HTS) specific components such as the magnetic material [159]
materials has renewed interest in research and or certain performance parameters as the output power
development of superconducting transformers. The [160], the load loss minimization [161,162] or the no-load
principal advantages of HTS transformers are: much lower loss minimization [163].
winding material content and losses, higher overload Apart from the transformer manufacturing cost,
capacity and possibility of coreless design. Some another criterion used for transformer evaluation and
considerations from design point of view are discussed in optimization is the Total Owing Cost (TOC) taking into
[141,142], while in [143] new perspectives of HTS account the cost of purchase as well as the cost of energy
transformer design are introduced. The development of losses throughout the transformer lifetime [164]. An
technology based on liquid nitrogen at temperature up to important part of the transformer cost optimization
79oK has reduced the complexity and cost of the research is devoted to the TOC minimization, as follows.
superconducting transformers [144,145]. A development Distribution transformer TOC optimization is analysed in
of three-phase 100 kVA superconducting transformer with [165-168]. Since the load losses are directly linked to the
amorphous core has been reported in [146]. A high- type of the considered load and the specific details of the
 The state of the art in engineering methods for transformer design and optimization: a survey 1155

network at the transformer installation point, a number of [9] M. Enokizono, T. Todaka, K. Nakamura, Journal of
versatile factors should be incorporated in the TOC Magnetism and Magnetic Materials 160, 61 (1996).
analysis. Such an analysis is performed in depth in [10] C. H. Yu, A. Basak, IEEE Transactions on Magnetics
[169,170]. 29(2), 1446 (1993).
Another aspect of transformer design optimization [11] A. Basak, C.-H. Yu, G. Lloyd, IEEE Transactions on
consists in providing design solutions in order to maintain Magnetics, 30(5), 3725 (1994).
certain aspects of transformer performance within the [12] M. Enokizono, N. Soda, IEEE Transactions on
limits imposed by the technical specifications. In this Magnetics 33(5), 4110 (1997).
context, the maintenance of short-circuit impedance and [13] C. Lin, C. Xiang, Z. Yanlu, C. Zhingwang,
losses within the acceptable tolerance is often addressed, Z. Guoqiang, Z. Yinhan, “Losses Calculation in
as in [171,172] and [173,174], respectively. Transformer Tie Plate using the Finite Element
Method,” IEEE Transactions on Magnetics
34(5), 3644 (1998).
5. Conclusion [14] D. A. Koppikar, S. V. Kulkarni, P. N. Srinivas,
S. A. Khaparde, R. Jain, “Evaluation of Flitch Plate
In the present paper, an overview of the literature Losses in Power Transformers,” IEEE Transactions
concerning transformer design has been undertaken, on Power Delivery 14(3), 996 (1999).
focusing on the progress realized in the past two decades. [15] G. F. Mechler, R. S. Girgis, IEEE Transactions on
Relevant publications from international journals have Power Delivery, 15(1), 198 (2000).
been selected, covering a broad range of engineering [16] Ed. G. teNyenhuis, G. F. Mechler, R. S. Girgis, IEEE
methods and design considerations. The difficulties to Transactions on Power Delivery 15(1), 204 (2000).
include and categorize the majority of the research in such [17] G. F. Mechler, R. S. Girgis, IEEE Transactions on
a vast field were overcome by a convenient survey Power Delivery 13(2), 532 (1998).
structure, taking into account various design [18] G. W. Swift, IEEE Transactions on Magnetics,
considerations. This survey provides important MAG-11(1), 61 (1975).
information on the main directions of the considered [19] S. Charap, F. Judd, IEEE Transactions on Magnetics,
research and the future trends in the field of transformer 10(3), 678 (1974).
design. [20] Z. Valkovic, IEEE Transactions on Magnetics,
MAG-18(2), 801 (1982).
Acknowledgment [21] F. Loffler, T. Booth, H. Pfutzner, C. Bengtsson,
K. Gramm, IEE Proceedings, Electric Power
This paper is part of the 03ED045 research project Applications, vol. 142(6), 371 (1995).
that is co-financed by E.U.-European Social Fund (75%) [22] Ed G. teNyenhuis, R. S. Girgis, G. F. Mechler, IEEE
and the Greek Ministry of Development-GSRT (25%). Transactions on Power Delivery 16(4), 648 (2001).
[23] A. Ilo, IEEE Power Engineering Review 22(3), 43
(2002).
References [24] A. Mae, K. Harada, Y. Ishihara, T. Todaka, IEEE
Transactions on Magnetics, 38(2), 829 (2002).
[1] R. M. Del Vecchio, B. Poulin, P. T. Feghali, [25] A. J. Moses, B Thomas, IEEE Transactions on
D. M. Shah, R. Ahuja, Transformer Design Magnetics, 10(2), 148 (1974).
Principles-With Application to Core-Form Power [26] M. Elleuch, M. Poloujadoff, IEEE Transactions on
Transformers, Gordon and Breach Science Magnetics, 34(5), 3701 (1998).
Publishers, Canada, 2001. [27] C. Nussbaum, H. Pfützner, T. Booth,
[2] S. V. Kulkarni, S. A. Khaparde, Transformer N. Baumgartinger, A. Ilo, M. Clabian, IEEE
Engineering Design & Practice, Marcel Dekker, Inc., Transactions on Magnetics, 36(1), 313 (2000).
New York, 2004. [28] P. S. Georgilakis, N. D. Hatziargyriou,
[3] IEEE Standard Terminology for Power and N. D. Doulamis, A. D. Doulamis, S. D. Kollias,
Distribution Transformers, IEEE Std Neurocomputing 23, 15 (1998).
C57.12.80-2002. [29] P. Georgilakis, N. Hatziargyriou, D. Paparigas,
[4] Use and Maintenance of ELVIM Oil-immersed S. Elefsiniotis, Journal of Materials Processing
Distribution Transformers, Schneider Electric AE, Technology 108, 209 (2001).
Greece, 2001. [30] P. S. Georgilakis, N. D. Hatziargyriou,
[5] J. H. Harlow, Electric Power Transformer A. D. Doulamis, N. D. Doulamis, S. D. Kollias, in
Engineering, CRC Press, New York, 2004. Proc. of the 21st 1999 IEEE International Conference
[6] D. Pavlik, D. C. Johnson, R. S. Girgis, IEEE on Power Industry Computer Applications, pp. 301-
Transactions on Power Delivery 8(1), 239 (1993). 308, May 1999.
[7] S. A. Holland, G. P. O'Connell, L.Haydock, IEEE [31] P. Georgilakis, N. Hatziargyriou, D. Paparigas, IEEE
Transactions on Magnetics.28(2), 1355 (1992). Computer Applications in Power, 12(4), 41 (1999).
[8] M. Enokizono, N. Soda, Journal of Magnetism and [32] M. Elleuch, M. Poloujadoff, IEEE Transactions on
Magnetic Materials, 196, 910 (1999). Magnetics, 39(2), 973 (2003).
1156 Eleftherios I. Amoiralis, Marina A. Tsili, Pavlos S. Georgilakis

[33] J. Proussalidis, N. Hatziargyriou, A. Kladas, Journal Magnetics, 28(2), 1344 (1992).

of Materials Processing Technology, 108, 217 (2001). [59] M. A. Tsili, A. G. Kladas, P. S. Georgilakis, A. T.
[34] A. A. Adly, IEEE Transactions on Magnetics, 31(6), Souflaris, D. G. Paparigas, Electric Power Systems
4253 (1995). Research, 76, 729 (2006).
[35] H. Akçay, D. Gökhan Ece, IEEE Transactions on [60] K. Zakrewski, B. Tomczuk, IEEE Transactions on
Power Delivery 18(2), 487 (2003). Magnetics, 32(3), 1637 (1996).
[36] R. Arseneau, E. So, E. Hanique, IEEE Transactions [61] C. A. Brebbia, R. Magureanu, Engineering Analysis
on Instrument and Measurement 54(2), 503 (2005. 4(4), 178 (1987).
[37] A. J. Moses, IEEE Transactions on Magnetics, 34(4), [62] M. A. Tsili, A. G. Kladas, P. S. Georgilakis, A. T.
1186 (1998). Souflaris, C. P. Pitsilis, J. A. Bakopoulos,
[38] R. S. Girgis, Ed G. teNijenhuis, K. Gramm, J.-E. D. G. Paparigas, IEEE Transactions on Magnetics,
Wrethag, IEEE Transactions on Power Delivery 40(2), 842 (2004).
13(2), 526 (1998). [63] M. Tsili, A. Kladas, P. Georgilakis, A. Souflaris, D.
[39] M. Albach, T. Durbaum, A. Brockmeyer, Proc. of Paparigas, Journal of Materials Processing
27th Annual IEEE Power Electronics Specialists Technology 161, 320 (2005).
Conference 2, (23-27), 1463 (1996). [64] J. Turowski, M. Turowski, M. Kopec, IEEE
[40] M. Dolinar, D. Dolinar, G. Stumberger, B. Polajzer, Transactions on Magnetics 26(5), 2911 (1990).
J. Ritonja, IEEE Transactions on Magnetics [65] D. J. Wilcox, W. G. Hurley, M. Conlon, IEE
42(10), 2849 (2006). Proceedings, Generation, Transmission and
[41] J. Pedra, L. Sainz, F. Córcoles, R. Lopez, M. Salichs, Distribution, 136(5), 308 (1989).
IEEE Transactions on Power Delivery [66] P. Raitsios, Journal of Materials Processing
19(1), 200 (2004). Technology 108, 246 (2001).
[42] N. Stranges, R. D. Findlay, IEEE Transactions on [67] S. H. Thilagar, G. S. Rao, Engineering Applications
Magnetics 36(5), 3457 (2000). of Artificial Intelligence 15(5), 429 (2002).
[43] R. Findlay, R. Belmans, D. Mayo, IEEE Transactions [68] J. H. Brunke, K. J. Fröhlich, IEEE Transactions on
on Magnetics, 26(5), 1990 (1990). Power Delivery 16(2), 276 (2001).
[44] D. Davies, A. J.Moses, IEEE Transactions on [69] M. Steurer, K. Fröhlich, IEEE Transactions on Power
Magnetics 20(4), 559 (1984). Delivery, 17(1), 155 (2002).
[45] P. Marketos, T. Meydan, IEEE Transactions on [70] L. Sáinz, F. Córcoles, J. Pedra, L. Guasch, IEEE
Magnetics 42(10), 2821 (2006). Transactions on Power Delivery 22(2), 986 (2007).
[46] S. V. Kulkarni, S. A. Khaparde, Proc. of IEEE Power [71] S.-D. Chen, R.-L. Lin, C.-K. Cheng, IEEE
Engineering Society Winter Meeting 3, 2269 (2000). Transactions on Power Delivery 20(3), 1947 (2005).
[47] A. Krawczyk, J. Turowski, IEEE Transactions on [72] S.G. Abdulsalam, W. Xu, V. Dinavahi, IEE
Magnetics 23(5), 3032 (1987). Proceedings, Generation, Transmission and
[48] S. Wang, S-C. Ji, Y. Li, Proc. of International Distribution 152(3), 328 (2005).
Conference on Power System Technology [73] C. E. Lin, C. L. Cheng, C. L. Huang, J. C. Yeh, IEEE
3, 1392 (2002). Transactions on Power Delivery 8(1), 246 (1993).
[49] D. A. Koppikar, S. V. Kulkarni, G. Ghosh, [74] A. A. Adly, IEEE Transactions on Magnetics, 37(4),
S. M. Ainapure, J. S. Bhavsar, IEE Proceedings 2855 (2001).
Science, Measurement & Technology [75] S. Salon, B. LaMattina, K. Sivasubramaniam,
145(4), 136 (1998). “Comparison of Assumptions in Computation of
[50] J.-F. Pern, S.-N. Yeh, IEE Proceedings, Science, Short Circuit Forces in Transformers,” IEEE
Measurement and Technology, 142, (3), 231 (1995). Transactions on Magnetics, 36(5), 3521 (2000).
[51] B. S. Ram, Proc. IEE Generation, Transmission and [76] H. Wang, K. L. Butler, IEEE Transactions on Power
Distribution, 145(6), 709 (1998). Delivery, 17(2), 500 (2002).
[52] T. Renyuan, L. Yan, L. Feng, T. Lijian, IEEE [77] T. Renyuan, L. Yan, L. Dake, T. Lijian, IEEE
Transactions on Magnetics, 32(3, 1641 (1996). Transactions on Magnetics, 28(2), 1418 (1992).
[53] R. Escarela-Perez, S. V. Kulkarni, N. K. Kodela, J. C. [78] S. L. Ho, Y. Li, H. C. Wong, S. H. Wang, R. Y. Tang,
Olivares-Galvan, IEEE Transactions on Power IEEE Transactions on Magnetics, 40(2), 687 (2004).
Delivery 22,(3), 1566 (2007). [79] N. Y. Abed, O. A. Mohammed, “Modeling and
[54] O. W. Andersen, IEEE Transactions on Power Characterization of Transformers Internal Faults
Apparatus and Systems, PAS-92(2), 682 (1973). Using Finite Element and Discrete Wavelet
[55] X. Cui, G. Zhang, IEEE Transactions on Magnetics, Transforms,” IEEE Transactions on Magnetics 43(4),
36(4), 1822 (2000). 1425 (2007).
[56] A.G. Kladas, M.P. Papadopoulos, J.A. Tegopoulos, [80] K. Zakrzewski, B. Tomczuk, D. Koteras, IEEE
IEEE Transactions on Magnetics Transactions on Magnetics 38(2), 1153 (2002).
30(5/2), 3487 (1994). [81] G. B. Kumbhar, S. V. Kulkarni, IEEE Transactions
[57] M. Djurovic, C. J. Carpenter, IEEE Transactions on on Power Delivery 22(2), 936 (2007).
Magnetics, 11(5), 1535 (1975). [82] T. Yanada, S. Minowa, O. Ichinokura, S. Kikuchi,
[58] K. Zakrewski, M. Kukaniszyn, IEEE Transactions on IEEE Transactions on Magnetics 34(4), 1351 (1998).
 The state of the art in engineering methods for transformer design and optimization: a survey 1157

[83] W. W. L. Keerthipala, Z. RuJing, T. Eu, L. Chionh, Computation 3, 2745 (2005).

C. Jinn, Proceedings of Power Electronic Drives and [108] Q. He, J. Si, D. J. Tylavsky, IEEE Transactions on
Energy Systems for Industrial Growth 1, 243 (1998). Power Delivery 15(4), 1205 (2000).
[84] C. Teoh, K. Soh, R. Zhou, D. Tien, V. Chan, “Active [109] L. Pierce, IEEE Transactions on Power Delivery
Noise Control of Transformer Noise,” in Proc. of 7(2), 920 (1992).
International Conference on Energy Management and [110] L. Pierce, IEEE Transactions on Power Delivery,
Power Delivery 2, 747 (1998). 7(3), 1347 (1992).
[85] M. Rausch, M. Kaltenbacher, H. Landes, R. Lerch, J. [111] L. Pierce, IEEE Transactions on Industry
Anger, J. Gerth, Journal of Sound and Vibration Applications 30(1), 170 (1994).
250(2), 323 (2002). [112] T. O. Rouse, IEEE Electrical Insulation Magazine
[86] E. Lesniewska, IEEE Transactions on Magnetics, 14(3), 6 (1998).
38(2), 1233 (2002). [113] Y. Bertrand, L.C. Hoang, Proceedings of the 7th
[87] D. D. Mairs, D. L. Stuehm, B. A. Mork, IEEE International Conference on Properties and
Transactions on Industry Applications, 25(2), 366 Applications of Dielectric Materials 2, 491 (2003).
(1989). [114] Caixin Sun, Jian Li, Xiaohu Li, S. Grzybowski,
[88] M. Hori, M. Nishioka, Y. Ikeda, K. Noguchi, K. IEEE Conference on Electrical Insulation and
Kajimura, H. Motoyama, T. Kawamura, IEEE Dielectric Phenomena, 680 (2006).
Transactions on Power Delivery 21(3), 1600 (2006). [115] D. Martin, Z. D. Wang, A. W. Darwin, I. James,
[89] M. Hori, M. Mitani, H. Sakurano, H. Motoyama, T. Proc. of IEEE Conference on Electrical Insulation
Kawamura, IEEE Transactions on Power Delivery and Dielectric Phenomena, 493 (2006).
22(1), 316 (2007). [116] C. P. McShane, IEEE Industry Applications
[90] M. Popov, L. van der Sluis, IEEE Transactions on Magazine 8(3), 34 (2002).
Power Delivery 16(3), 401 (2001). [117] T. V. Oommen, IEEE Electrical Insulation Magazine
[91] T. K. Saha, P. Purkait, F. Müller, IEEE Transactions 8(1), 6 (2002).
on Power Delivery 20(1), 149 (2005). [118] C. Perrier, A. Beroual, J.-L. Bessede, IEEE
[92] T. K. Saha, IEEE Transactions on Power Delivery Transactions on Dielectrics and Electrical Insulation
18(4), 1293 (2003). 13(3), 556 (2006).
[93] P. M. Balma, R. C. Degeneff, H. R. Moore, L. B. [119] C. Perrier, A. Beroual, J.-L. Bessede, IEEE
Wagenaar, IEEE Transactions on Power Delivery International Symposium on Electrical Insulation,
14(3), 960 (1999). 237 (2004).
[94] L. Pierce, IEEE Transactions on Power Delivery 9(2), [120] C. P. McShane, J. Luksich, K. J. Rapp, Proc. of
1160 (1994). IEEE-IAS/PCA Cement Industry Technical
[95] Ed G. teNyenhuis, R. S. Girgis, G. F. Mechler, Conference, 141 (2003).
G. Zhou, IEEE Transactions on Power Delivery [121] P. Picher, L. Bolduc, A. Dutil, V. Q. Pham, IEEE
17(4), 991 (2002). Transactions on Power Delivery 12,(1), 257 (1997).
[96] K. Preis, O. Bíró, G. Buchgraber, I. Ticar, IEEE [122] O. Bíró, S. Außerhofer, G. Buchgraber, K. Preis, W.
Transactions on Magnetics 42(4), 999 (2006). Seitlinger, IET Science, Measurement &
[97] C. C. Hwang, P. H. Tang, Y. H. Jiang, IEE Technology 1(1), 2 (2007).
Proceedings, Electric Power Applications, 152(4), 832 [123] S. Lu, Y. Liu, IEEE Transactions on Power
(2005). Delivery, 8(3), 1367 (1993).
[98] S. A. Ryder, I. J. Vaughan, IEEE Transactions on [124] W. C. Viana, R. J. Micaleff, S. Young, F. P.
Power Delivery 19(2), 637 (2004). Dawson, IEEE Transactions on Magnetics, 35(5),
[99] G. Swift, T. S. Molinski, W. Lehn, IEEE Transactions 3532 (1999).
on Power Delivery 16(2), 171 (2001). [125] E.F. Fuchs, Y. You, D.J. Roesler, IEEE Transactions
[100] G. Swift, T. S. Molinski, R. Bray, R. Menzies, IEEE on Power Delivery 14(2), 443 (1999).
Transactions on Power Delivery 16(2), 176 (2001). [126] A. Ahfock, A.J. Hewitt, IEE Proceedings, Electric
[101] Z. Radakovic, K. Feser, IEEE Transactions on Power Applications, 153(4), 601 (2006).
Power Delivery 18(4), 1284 (2003). [127] A. R. Garcez Castro, V. Miranda, IEEE Transactions
[102] W. H. Tang, Q. H. Wu, Z. J. Richardson, IEEE on Power Systems 20(2), 717 (2005).
Transactions on Power Delivery 19(3), 1112 (2004). [128] X. Hao, S. Cai-xin, IEEE Transactions on Power
[103] V. Galdi, L. Ippolito, A. Piccolo, A. Vaccaro, Delivery 22(2), 930 (2007).
Electric Power Systems Research 60, 107 (2001). [129] J. L. Guardado, J. L. Naredo, P. Moreno, C. R.
[104] D. Susa, M. Lehtonen, H. Nordman, IEEE Fuerte, IEEE Transactions on Power Delivery 16(4),
Transactions on Power Delivery, 20(1), 197 (2005). 643 (2001).
[105] D. Susa, M. Lehtonen, IEEE Transactions on Power [130] A. S. Farag, M. Mohandes, A. Al-Shaikh, IEEE
Delivery, 21(4), 1961 (2006). Transactions on Power Delivery 16(4), 631 (2001).
[106] D. Susa, M. Lehtonen, IEEE Transactions on Power [131] Y. C. Huang, IEEE Transactions on Power Delivery
Delivery 21(4), 1971 (2006). 18(3), 843 (2003).
[107] W. H. Tang, E. Prempain, Q. H. Wu, J. Fitch, Proc. [132] J. K. Pylvänäinen, K. Nousiainen, P. Verho, IEEE
of the 2005 IEEE Congress on Evolutionary Transactions on Power Delivery 22(1), 201 (2007).
1158 Eleftherios I. Amoiralis, Marina A. Tsili, Pavlos S. Georgilakis

[133] P. S. Georgilakis, J. A. Katsigiannis, K. P. 5th International Conference on Electrical Machines

Valavanis, A. T. Souflaris, Journal of Intelligent and and Systems, ICEMS 2001, 1, 242 (2001).
Robotic Systems, 45(2), 181 (2006). [154] S. Elia, G. Fabbri, E. Nistico, E. Santini,
[134] W. H. Tang, S. He, Q. H. Wu, Z. J. Richardson, International Symposium on Power Electronics,
International Journal of Innovations in Energy Electrical Drives, Automation and Motion,
Systems and Power, 1(1), 46 (2006). SPEEDAM 2006, 1473 (2006).
[135] Y. Hong-Tzer, L. Chiung-Chou, IEEE Transactions [155] N. Tutkun, A. Moses, Journal on Magnetism and
on Power Delivery 14(4), 1342 (1999). Magnetic Materials 277(1-2), 216 (2004).
[136] T. K. Saha, P. Purkait, IEEE Transactions on Power [156] E. I. Amoiralis, P. S. Georgilakis, T. D. Kefalas, M.
Delivery 19(3), 1127 (2004). A. Tsili, A. G. Kladas, IEEE Transactions on
[137] S. Sieradzki, R. Rygal, M. Soinski, IEEE Magnetics, 43(4), 1633 (2007).
Transactions on Magnetics, 34(4), 1189 (1998). [157] L. H. Geromel, C.R. Souza, IEEE Canadian
[138] M. Yamamoto, T. Mori, T. Kawasaki, K. Tsutsui, T. Conference on Electrical and Computer Engineering
Itoh, T. Yasigawa, IEEE Transactions on Magnetics 1, 285 (2002).
20(5), 1771 (1984). [158] R. A. Jabr, IEEE Transactions on Magnetics 41(11),
[139] A. Coombs, M. Lindenmo, D. Snell, D. Power, 4261 (2005).
IEEE Transactions on Magnetics 37(1), 544 (2001). [159] C. J. Wu, F. C. Lee, IEEE Transactions on
[140] T. Matsuura, K. Naqayama, S. Hagiwara, M. Higaki, Magnetics 16(5), 755 (1980).
IEEE Transactions on Magnetics 26(5), 1993 [160] O. W. Andersen, IEEE Computer Applications in
(1990). Power, 4(1), 11 (1991).
[141] T. Nitta, K. Misawa, H. Nomura, IEEE Transactions [161] N. D. Doulamis, A. D. Doulamis, P. S. Georgilakis,
on Magnetics 32(4), 2381 (1996). S. D. Kollias, N. D. Hatziargyriou, Integrated
[142] E. N. Andreev, Li. I. Chubraeva, Journal of Computer-Aided Engineering 9, 37 (2002).
Materials Processing Technology 181(1-3), 25 [162] T. H. Pham, S. J. Salon, S. R. H. Hoole,
(2007). Transactions on Magnetics 32(5), 4287 (1996).
[143] M. Ariante, A. Formisano, F. Marignetti, R. [163] N. D. Doulamis, A. D. Doulamis, Proc. of IEEE
Martone, G. Masullo, A. Matrone, R. Quarantiello, International Conference on Systems, Man and
G. Rubinacci, F. Sangiorgi, M. Scarano, S. Silvestri, Cybernetics 5, 5 (2002).
F. Villone, M. Zigon, Journal of Physics: [164] B. Kennedy, Energy Effcient Transformers,
Conference Series 43, 845 (2006). McGraw-Hill, 1998.
[144] J. K. Sykulski, C. Beduz, R. L. Stoll, M. R. Harris, [165] J. F. Baranowski, T. A. Edison, P. J. Hopkinson,
K. F. Goddard, Y. Yang, IEE Proceedings, Electric IEEE Transactions on Power Delivery 7(2), 614
Power Applications 146(1), 41 (1999). (1992).
[145] J.-M. Joung, S.-M. Baek, S.-H. Kim, IEEE [166] S. Y. Merritt, S. D. Chaitkin, IEEE Industry
Transactions on Applied Superconductivity 14(2), Applications Magazine 9(6), 21 (2003).
928 (2004). [167] S. Y. Merritt, S. D. Chaitkin, IEEE Industry
[146] J.-K. Lee, W.-S. Kim, S.-Y. Hahn, IEEE Applications Magazine 11(4), 66 (2005).
Transactions on Applied Superconductivity 9(2), [168] P. S. Georgilakis, Journal of Materials Processing
1293 (1999). Technology 181, 307 (2007).
[147] H. Kamijo, H. Hata, H. Fujimoto, K. Ikeda, T. Herai, [169] D. L. Nickel, H. R. Braunstein, IEEE Trans. Power
K. Sakaki, H. Yamada, Y. Sanuki, S. Yoshida, Y. Apparatus and Systems 100(2), 788 (1981).
Kamioka, M. Iwakuma, K. Funaki, IEEE [170] D. L. Nickel, H. R. Braunstein, IEEE Trans. Power
Transactions on Applied Superconductivity 15(2), Apparatus and Systems 100(2), 798 (1981).
1875 (2005). [171] N. Takahashi, T. Kitamura, M. Horii, J. Takehara,
[148] K. Fukuoka, M. Hashimoto, IEEE Transactions on IEEE Transactions on Magnetics 36(4), 1089
Applied Superconductivity 14(2), 1126 (2004). (2000).
[149] J-T Kim, W.-S. Kim, S.-H. Kim, K.-D. Choi, J.-H. [172] M. A. Tsili, A. G. Kladas, P. S. Georgilakis, A. T.
Han, G.-W. Hong, S.-Y. Hahn, IEEE Transactions Souflaris, D. G. Paparigas, IEEE Transactions on
on Magnetics, 41(5), 1888 (2005). Magnetics 41, (5), 1776 (2005).
[150] A. Takizawa, Y. Ono, S. Inaka, K. Konno, Y. [173] J. C. Olivares, Y. L. Jose, M. Canedo, R. Escarela-
Nozaki, T. Suzuki, S. Yamazaki, S. Saito, T. Pérez, J. Driesen, P. Moreno, IEEE Transactions on
Shirone, IEEE Power Engineering Society 1999 Power Delivery, 18(3), 821 (2003).
Winter Meeting, 2, 1036 (1999). [174] M. Rizzo, A. Savini, J. Turowski, Proc. Fifth
[151] S. Trabulus, Proc. of 1999 Annual Report International Conference on Electrical Machines and
Conference on Electrical Insulation and Dielectric Systems 1, 182 (2001).
Phenomena (1), 321 (1999).
[152] P. S. Georgilakis, M. A. Tsili, A. T. Souflaris, _______________________________
Journal of Materials Processing Technology 181, *
Corresponding author: [email protected], [email protected]
260 (2007).
[153] L. Hui, H. Li, H. Bei, Y. Shunchang, Proceedings