See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/272853347

Computer Simulation of Theoretical Model of Electromagnetic Transient

Processes in Power Transformers

Article  in  International Journal of Information Technology and Computer Science · January 2014

DOI: 10.5815/ijitcs.2014.01.01

CITATIONS READS

8 109

2 authors:

Slobodan Bjelic Zorica Bogicevic

University of Pristina 12 PUBLICATIONS   32 CITATIONS   
55 PUBLICATIONS   107 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Erasmus + Project - "Creating the Network of Knowledge Labs for Sustainable and Resilient Environments" [KLABS] [561675-EPP - 1-2015-1-XK-EPPKA2-CBHE-JP] View
project

AI - Based Method And Mathematical Model For Analysis Of Current And Voltage In Elements Of Power Systems, ID 143934. View project

All content following this page was uploaded by Slobodan Bjelic on 03 May 2016.

The user has requested enhancement of the downloaded file.

I.J. Information Technology and Computer Science, 2014, 01, 1-12
Published Online December 2013 in MECS (http://www.mecs-press.org/)
DOI: 10.5815/ijitcs.2014.01.01

Computer Simulation of Theoretical Model of

Electromagnetic Transient Processes in Power
Transformers
Slobodan Bjelić
Faculty of Technical Sciences, Kosovska Mitrovica, Serbia
E-mail: slobodanbjelić49@yahoo.com

Zorica Bogićević
Advanced Polytechnic School of Professional Sciences, Zvecan, Serbia
E-mail: zocabogi@gmail.com

Abstract — The paper is presenting theoretical
analytical model and computer simulation of
electromagnetic transient process in a transformer.
Transformer parameters in a selected transitional
process have been analyzed. Theoretical model refers to
an energetic transformer with concentrated parameters
with consideration of parameters of mutual inductance
M. Simulation was performed on a personal computer
using the software program MATLAB SIMULINK. The
computer simulation confirmed the possibility of
transitional process analysis in transformer’s windings
with concentrated parameters.
Index Terms— Computer, Simulation, Transient
Process, Model, Transformer, Concentrated Parameters
I. Introduction
Transformer’s load causes a change in voltage. This
change is usually manifested as a decrease (down) of
voltage relative to the idle. The transformer should be
designed to withstand the possible tension throughout
its life. This paper describes changes in the transformer
during load changes through computer simulations that
indicate a failure or defect.
Computer simulation today presents one of the
leading methods for solving, describing, understanding
and analysis of complex dynamic systems in the fields
of technical sciences.
The MATLAB software is designed for solving
various mathematical and engineering problems Which
were modeled by the application of linear algebra and a
master computer where this simulation has been
performed, and that is why the simulation was
performed precisely in this software package.
Mathematical model of power transformer is
completely derived from the literature [1], and is based
on the model derived in the software package
MATLAB and Simulink[7]. The simulation’s goal of the
set theoretical models is to obtain relevant information
about the behavior of the power transformer in the
transition processes.
In the first example of a transformer of high power
and a second example of a transformer of low power it
has been shown that such a computer simulation in a
qualitative way can describe the transient processes and
by doing so a better analysis of their behavior can be
performed.
The remainder of this paper is organized as follows:
Section 2 gives a theoretical model of the transition
process in the ideal and the real transformer with linear
magnetic characteristics. Section 3 describes dynamic
changes in real power transformer in MATLAB
Simulink. It describes two simulations: Simulation 1-
Simulates idling of transformer. Secondary current is
equal to zero. Simulation was performed for the real
three-phase power transformer SIEMENS - ONAF /
ONAN.
Simulation 2- Simulates a short circuit and controls a
secondary current set to the rated current for the real
three-phase power transformer SIEMENS - ONAF.
Graphs fluxes, current magnetization and voltages on
the primary and the secondary are given at the end of
the simulation process. Conclusion and future work are
presented in the final section.
II. Theoretical Model of Transient Process in
Transformer
The transformer consists of ferromagnetic circuit and
operates as electromagnetic connection of two electric
windings: 1. primary is connected to the source of
alternating current and 2. the secondary winding to
which electric loading is connected. The task of the
magnetic circuits is to create spatial distribution of

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
2 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers
magnetic excitation forces and magnetic fluxes which
produce currents in winding at a circuit or strange
magnetic fields[6]. The processes in a three-phase
transformer are equivalent to processes in single-phase
one if the influences of the other two phases are
considered according to assumptions as in the model of
an ideal and symmetric three-phase transformer, where
according to figure 1:
Fig. 1: a)
Fig. 1: b)
Fig. 1: a) Three separated magnetic circuits and b) equivalent scheme
of transformer with concentrated parameters
1. Conductance of magnetic circuits for all three phases
are equal: a  b  c  
2. Sum of magnetic excitation forces of three phases
   c) due to the dissipation of magnetic fluxes between
equals to zero M a  M b  M c  0
Copyright © 2014 MECS

3. According to IV Maxwell equation divB  0 ,
 a   b   c  0
Currents and voltages in windings can oscillate
propagation through coils if the winding includes; The
theory of wave spreading through coil’s threads covers
parameters of capacitance towards mass, winding
inductance, capacitances between adjacent coils, mutual
inductivity of one thread to another, reaction of primary
coil on secondary and vice versa, turbulent currents in
magnetic core , losses caused by hysteresis.
In the model, values of primary and secondary
winding are R p  R1[] , Rs  R2 [] and values of
inductance are L p  L1 [ H ] , Ls  L2 [ H ]
Fig. 2: a) One equivalent scheme of three-phase transformer, with
Hysteresis curve
Nonlinear characteristic of magnet biasing, figure 2 is
dependence B  f ( H Fe ) which has an hysteresis
form of a distorted rectangle and high values of
magnetic penetrability   dB / dH .
High values of magnetic penetrability   dB / dH
help to minimize influences of air gaps and magnetic
contours on which magnetic fluxes close.
Magnetic induction and strength of the fields B , H
have different values in different points of cross section
of the core. Values are variable due to three reasons:
a) due to the value change of cross section
b) due to the heterogeneity of magnetic domains
(ferromagnetic, gap filled with air, etc), and
adjacent coils.
I.J. Information Technology and Computer Science, 2014, 01, 1-12
 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers 3

Magnetic characteristics of magnetic circuits of Differences in the process of magnetic biasing, which
devices such are transformer, dependence Bsr  f (H ) , occur on hysteresis curve.
differs from magnetic characteristics of material Magnetic parameters must be determined on a
B  f (H ) due to following reasons [1]: finished magnetic circuit which has its own geometric
shape and winding position.
 Lengths of contours where magnetic fluxes close are
not equal, and there are also fluxes of dissipation. According to the Faraday law, the electro motor force
Influence of unequal lengths of contours can be of primary and secondary winding, u1 and u2 are
lessened only by assumption that a magnetic circuit is consequences of changes of fluxes:
composed of elementary parts with the same
magnetic characteristics which are packed in order in t t
1 1
the direction of a planned contour.  
Np 0
u p dt    u1dt ,
N1 0
(4)
 Inequalities of cross sections along contours of fluxes.
In calculations that influence, this can be lessened t t
1 1
with the usage of the smallest value of a cross section
S m along contours of the magnetic circuit.
 
Ns 0
u s dt  
N 2 0
u 2 dt , (5)

 Influence of connections, cracks and presence of

magnetic gaps in the circuit filled with air along 2.1 Theoretical Model Of A Transitional Process in
contours of flux closures. According to the second of an Ideal Transformer with Linear
Kirchhoff’s laws on magnetic circuits [2]: Magnetization Characteristic

  B
H l
j
j j  hFe  l 
0 , (1) u iz . p  U iz .m sin(  t   ) (6)
 MPS  H (l   )
Voltage equations are:
hFe is the intensity of the field which acts as the
magnetic core, B  b  B - is the inductance value if
di p dis
uiz . p  L p M 0, (7)
there is no dissipation in the gap on the cross section, dt dt
H - mean value or real value of field intensity which
acts on the cross section , H p - value of field intensity dis di p di M di p
Ls M 0 s  (8)
which would exist on the cross section if gaps would dt dt dt Ls dt
not exist.

MPS F  B Connection coefficient k c2  M has the value

H   hFe   hFe  H p , (2) L p Ls
l  l l 0
M
kc   1 if the fluxes include the winding of
where  e  B  0 H is the relative value of magnetic L p Ls
conductivity of a magnetic circuit,   B / H is the the primary and secondary are equal in value. In
absolute value of magnetic penetrability of material, approximate calculations, the following expressions
l apply:
 m p is the magnetic penetrability in the form of a
 N1 2 N
l Lp  0 r ( ) V , Ls   0  r ( 2 ) 2 V
magnetic  m p circuit. l l
 , (9)
N1 N 2
For magnetic circuits where
l
 mp 1 , l   M  L p Ls   0  r 2 V
 l
1 1 1 
applies    e  1.
e  1  1 N1 N 2
M  L p Ls   0  r V , (10)
This procedure leads to the expression for differential l2
magnetic penetrability: H
where  0  4  10 7 r is the magnetic
m ,
1 H 1 1 1
 ,   , (3) penetrability of the air and core, N 1 , N 2 are the
 de B  de  d m p
numbers of the primary and secondary coils and

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
4 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers

V volume as the part of the space where the windings Substituting the values for derivative (di s / dt ) in (6)
are placed, figure 3. In the primary transformer circuit
of the system of (9), the following is obtained:
active losses presented with resistance exist R p and in
the secondary circuit with resistance Rs , if the power M 2 di p N (14)
u p  L p (1  )  [ R p  ( 1 ) 2 Rs ]  i p
transformer is directly connected to voltage source: L p  Ls dt N2

di p di s From the first equation of the system of (6) is also:

u iz . p  L p M  Rp  ip , (11)
dt dt
dis 1 di p
 (uiz . p  R p i p  L p ) (15)
di di p  dt M dt
Ls s  M  R s i s  0 / , (12)
dt dt t When the primary transformer is connected to
external electric circuit with a voltage source
di s d 2i p d 2is uiz. p  U iz.m sin  t    and the parameters of the
Rs M  L s 0 (13)
dt dt 2 dt 2 circuit are Riz , Liz and the secondary transformer
winding is connected, loading with the parameters
In the secondary circuit of single-phase transformer, Rload , Lload in the primary transformer are: total
three typical regimes can occur. active resistance R p  R p  Riz , total inductance
a) Idle operation: L p  L p  Liz , and in the secondary transformer the

di p active resistance is Rs  Rs  Rload , and the total

Rs    i s  0  u s   M inductance Ls  Ls  Lload while time constants gain
dt
other values:
b) Short circuit:
dis di i L N R p R p  Riz
R s  0   Ls  M s 0 s  s  2  p   ;
dt dt i p M N1 L p Lp  Liz
, (16)
c) Loading on secondary is small value but is not R R  Rload
 s  s  s
equal to zero us  is Zload , Rs  Zload  0 : Ls Lp  Lload
L di M di p Rs
Rs  0  i s  s i p s    is
M dt Ls dt Ls

Fig. 3: Electromagnetic connection of transformer winding with concentrated parameters

With strongly paired magnetic circuits, as in the
transformer here, the following applies: M
With these substitutions which refer to a new
connection coefficient k c , effective inductance of the
primary Lef . p and the mutual inductance M and the
2
 k c2 L p Ls . introduction of time constants to the primary circuit
 p and the secondary circuit  s is:

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers 5

M2 In the moment of closing of both switches, figure 3

k c2  , Lef . p  (1  k c2 ) L p , (17) (in the circuit of primary and secondary) values are:
i p 0  0 for is 0  0 and u p 0  U iz.m sint  .
L p Ls

For which an homogenous part according to

convolution theorem  / t  j  p applies to the
From the (7) of the system we obtain the first
derivative:
equation:

 p s (1  kc 2 ) p 2  ( p   s ) p  1  0 , di p Ls

(18)  u p (0)
dt t 0
L p Ls  M 2
Roots of characteristic equation, when , (27)
uiz . p (0)
decrement D  0 , are: 
L p (1  kc2 )
 ( p   s )  ( p   s ) 2  4 p s (1  k c2 )
p1, 2  ,(19)

2 p s 1  k c2  Integrating constants are determined from the
boundary conditions:
and the decrement of the equation is greater than zero,
the roots are multiple and different: A1 p  A2 p    I pm cos(   p )
uiz . p (0) , (28)
 p s (1  kc 2 ) p 2  ( p   s ) p  1  0 , (20) 
Lp (1  kc2 )
The current of the primary is the sum of a free
The value of the current i s is obtained from the
component i psl and a forced component i ppr of the
differential equation:
primary current:

d 2is di
i p  i psl  i ppr  A1e p1t  A2 e p2t  i ppr , (21)  p s (1  k ) 2  ( p   s ) s  is
2
c
dt dt
, (29)
The forced component is created by the voltage of the M duiz. p
 ( )
resource u iz. p  U m sin  t    as well as the Rp Rs dt
harmonic function:

i ppr  I pm sin   t     p ,  (22) For the homogenous part of differential equation the
same decrement applies:
If the expression for the forced component i ppr
substitutes in (19) values for I pm , then  p shall be  p s (1  kc 2 ) p 2  ( p   s ) p  1  0 , (30)
determined:
The solution for the secondary current is the sum of
Rising to the second power and then extracting of the free and the forced component:
root of the left and right side of the last equation is
obtained:
is  issl  ispr  B1e p1t  B2 e p2t  ispr , (31)

I pm [1   p s (1  k c2 ) 2 ] 2  ( p   s ) 2  2 
,(23) The forced component is defined by a time form:
U
 sin(  t     p   )  iz .m 1   s2  2 sin(  t     )
R p ispr  I sm sin(  t     s ) , (32)

( p   s ) ,   arctg ( s ) , (24) If the expression for the forced component i spr is
  arctg
1   p s (1  k c2 ) 2
substituted in for (34) the values for I sm i  s will be
( p   s ) determined:
 p  arctg  arctg ( s ) , (25)
1   p s (1  kc2 ) 2
MUiz .m 1
I sm 
I pm 
U iz .m 1  2
s
2
, (26) Rp Rs [ p s (1  kc ) 2 ]2  ( p   s )2  2 , (33)
2

R p [1   p s (1  k c2 ) 2 ]2  ( p   s ) 2  2

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
6 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers

( p   s )  uiz . p 0  p s

2

 s  arctg  , (34) 1
B1   cos2 (   p ) , (46)
1   p s (1  k c ) 2
2 2 2
I sm M  p   s
Integrated constants are obtained from two conditions:
1. At the moment of closing of both switches (on the The values of transformer’s voltage are calculated of
primary and the secondary) the value is is 0  0 , and 2. the system of (9):
from (7) the first derivative is:
dis
2
u p u 1  M , (47)
dis M di p k u c iz . p (0) dt
  , (35)
dt t 0
Ls dt t 0
M (1  k c2 ) di p
u s u 2  M , (48)
If there are no magnetic dissipations the values are: dt

1 In transformers of 10  Vn  110kV the voltages for

kc  1 , p1   , p2   , (36)
 p   s  basic harmonic f  50Hz values of relation of winding
parameters are such an error of 0,1% occurs:
i p  i psl  i ppr  A1e p1t  i ppr , (37)
X X  L  4 R 
is  issl  ispr  B1e p1t
 ispr , (38) L , (49)
  ( )  4  314  1256  1
R
For primary the values are:
1   s2 2  s
 , (50)
U m  p 1    s
2 2 1  ( p   s ) 2
 p   s
I pm  , (39)
L p 1  ( p   s ) 2
U iz .m  p s
I pm 
L p  p   s
Um  p I pm , (51)
 , (40)  p s I pm
L p 1  ( p   s ) 2 1   2 s2   L
 p   s U iz .m p
 p s uiz . p. (0)
A1   I pm sin(   p ) u iz . p (0)
 p   s L p , (41) A1  I pm [  sin(   p )] , (52)
U iz .m
 p  arctg [( p   s ) ]
uiz . p  0 
t

 p  s 
and for secondary the are values: i p  I pm {e [  sin(   p )]
U iz .m , (53)
M  s  sin(t     p )}
I sm  I pm , (42)
Ls 1   2 2
s

U iz .m  p s M M
 p s uiz . p (0) I sm   I pm , (54)
B1    I sm cos(   p ) , (43) L p ( p   s ) Ls Ls
 p   s M

 s   p   sin(   s )  cos(   p ) , (55)
 s  arctg [( p   s ) ]  , (44) 2
2
uiz . p (0) Lp M
1 uiz . p (0)  p s
2
B1  I pm [  cos(   p )] , (56)
A1   sin 2 (   p ) ,(45) U iz .m M Ls
I 2
pm L p  p   s

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers 7

In the idle mode: Simulation and analysis of the transformer’s behavior

in transient processes was preformed in idle operation
di p and is consisted of the described simulation models of
Rs   / Ls    i s  0  u s   M , (57)
dt voltages and source parameters and transformers along
with blocks which simulate voltages of the transformer.
U iz .m  p
I pm  , (58) In accordance with technical practice it is necessary
Lp 1  ( p )2 to determine resistance and inductance of windings in
the system per unit.
Base values are nominal output Sn VA , nominal
This value also presents the current of magnetization:
frequency f n Hz , and nominal voltages Vn V  and
t
 uiz . p (0) effective voltage value Vrms which corresponds to
 p
im  i10  I pm {e [  sin(   p )] winding.
U iz .m , (59)
The base unit resistance, and base and unit
 sin(t     p )} inductance which are used for each winding are:


t
[Vn ]2 Rbase
 sin    p ]
1  p uiz. p (0) Rbase  Lbase 
us  MI pm { e [ Sn 2f n
 p U iz .m ,(60) ,
R[] L[ H ]
  cos(t     p )} R( p.u )  L( p.u ) 
Rbase [] Lbase [ H ]
Value of flux from (5) is: The unit value of active resistance of the magnetic
t
circuit Rm is based on indicated power Sn [VA] and on
1
N 2 0
 u s dt  , (61) the nominal voltage of the winding 1.

In the short-circuit mode conditions are: III. Transients in Transformers in MATLAB

SIMULINK
dis di
Rs  0   Ls M s 0 If the process is simulated with a residual flux phi0,
dt dt , (62) the second point of a saturation characteristic is on the
is Ls coordinate and corresponds to zero value of the current,
  as shown on figure 4.b. The characteristic of saturation
ip M is input with ( i , phi ) values per unit in the system, and
with start from couple (0,0). The Power Block (PSB)
uiz . p  0 
t
 converts vectors of flux p.u. and vectors of the current
 s
i p  I pm {e [  sin(   p )] I pu into standard units which shall be used in the
U iz .m , (63)
model of saturation:
 sin(t     p )}

t
   pu base ,  pu   / base
Ls  s uiz . p (0)
is  I pm  sin(   p )] , (64) ,
M
{e [
U iz .m I  I pu Ibase , I pu  I / I base
 sin(t     p )}

where the base values of flux and current have values

which correspond to nominal voltage, power and
2.2 Calculation of The Transformer’s Para- Meters frequency.

The model of the transformer is accomplished, using
the software program MATLAB-SIMULINK in such a
way that its inputs are vectors of source voltage and
source parameters and windings of the transformer, and
outputs are vectors of flux, currents of magnetic biasing
and variable equations of voltage value of both primary
and secondary.
Nominal power and frequency: Nominal apparent
power Sn VA , and frequency f n Hz , of transformer.
Winding parameters 1: Nominal effective value of
voltage in V  , resistance and drive inductance of
winding 1 in p.u.. Winding parameters 2: Nominal
effective value of voltage in V  , resistance and drive
inductance of winding 2 in p.u.. The characteristic of

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
8 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers

saturation-denoted as a flow/diagram of current pair V2n 36,75

(p.u.)-flux (p.u.) which starts from the point (0,0). kt    0,2988  0,3
V1n 123
Active losses in the core of magnetic circuit and initial
flux: Active losses of the power in the core which are Reactance of the primary and secondary windings are:
included into the resistance parameter Rm in p.u.. For

 7.56
example: Denoting loss of active power as 0,2% at X k 15.13
nominal voltage nominal value for Rm  500 p.u. also X k1  
2 2
can be denoted as the initial flux phi0 (p.u). The initial
0,32  0,68
X k 2 15,13
flux has special importance in transformer feeding. If X k2  kt 
phi0 is not denoted, the initial flux is automatically set 2 2
for simulation of a steady-state start.
Dissipative inductances of primary and secondary are:
The following has been measured: Voltages of
windin U1  U p  U N 1 ( p.u.) , U 2  U s  U N 2 ( p.u.) ,
 0,024H 
X k1 7.56
L1  L 1  
Currents of winding: I N1 , I N 2 . Magnetizing  314
current: I mag . Flux:  .
 0.0022H 
X k2 0.68
L2  L 2  
Outputs and inputs: One input, one output, or three  314
outputs (if they exist) instantly have the same polarity.
If on the input, with the third winding equaling 0 is Impedance of magnetic biasing is:
found, then in the block-set it is implemented in the
transformer with two windings and magnetic circuit and V1nV1n V 21n
the display shows an icon which symbolizes a Zm  X m   
transformer. 3  i0 % I nV1n i0 S n
123210 6
Limitations: A winding on the icon can vary. A   30,26k
variable winding is internally and directly connected to 0,005  100  10 6
the resistance in the circuit, and an invisible connection
has no influence on voltage and current measurement. Inductance on the primary side is:
The flux saturation model does not include hysteresis,
figure 4.b. The circuit is available in the file 30.26  10 3
 96,36H 
Xm
psbxfosaturable.mdl file. Lm  
 314

Resistance of transformer winding (primary and

3.1 Simulation No. 1 secondary) is:
Simulation was performed on the real three-phase
1 Pcun 1 Pcun
power transformer SIMENS-ONAF/ONAN R1  R p  R1k   ( 3V1n ) 2 
(dimensions: length 7800 mm , width 3250 mm , height 2 3I1n 2 3  S n
2 2

6100 mm , transformer has fans which start to work at

1 Pcun 2 1 285  103
higher loadings; oil flow depends only on siphon effect)  V1n  1232106  0,2156
with the following information: connection Yy , current 2 S n2 2 100 21012
of idle operation i0 %  0,5% , voltage of short circuit
uk %  10,5% , voltages of primary and secondary
1 Pcun 1 Pcun
R2  Rs  R2 k   ( 3V2 n ) 2 
U np  U n1  71.09kV  , U ns  U n2  21,24kV  , 2 3I 22n 2 3  S n2
S n  100MVA , PFe  49kW , Pcun  285 kW ,   
1 Pcun 2
V2 n
Vnp  Vn1  123  110  8%kV  , Vns  Vn2  36,75kV  . 2 S n2
1 285 103
From the primary side, the reactance of the short  2 12
36, 752 106  0, 0194   
circuit is: 2 100 10

u % V12n 10 1232  10 6 Rbase 

Vn 2  50.43
Xk  k   15.13 Sn
100 S n 100 100  10 6
Rbase 50.43
The ratio of the number of coils of the primary and Lbase    0,1606
2f n 314
secondary windings is:

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers 9

123e3 / sqrt3 1.0 1.22], residual flux = 0.6 p.u. is presented as a

sqrt2  3,19Wb
V1
 base  2 partially linear curve of flux dependence from the
2f n 2  50
magnetization current

sqrt2  1983A
Sn 100e6 Input values which correspond to dialog box are:
I base  2
V1 123e3 / sqrt3
0.2156
R1   0.004275 p.u .
The three-phase transformer 100 MVA  , 50.43.
123/36.75 kV  is single-phase fed from the resource
50 Hz  . Transformer: Nominal power 100e6,50 Hz  , 0.024H 
L1   0,1494 p.u .
parameters of winding 1 (primary): 123e3 Vrms/sqrt(3), 0.1606H 
R  0,00427 p.u., L  0,145 p.u., parameters of winding
2 (secondary): 36.7e3. Active losses in the magnetic
circuit: 1000p.u. Characteristic of saturation: [0 0; 0 1.0;

Fig. 4: a) Simulation of idle operation of transformer and b) characteristics of block saturation of saturated transformer

In this simulation process, figure 4 which presents a 3.2 Simulation No. 2

regime of idle operation of a transformer shows the
influence of the characteristic of saturation and time
constants of a transformer. In that moment, the values
of the flux and magnetization current start to increase
which cause the increase of the voltage in the primary
and secondary, where the voltage on the secondary
transformer can have a higher unit value from the
voltage of primary transformer. During the process the
time constants of the primary and secondary as well as
the nonlinear characteristic of magnetic saturation have
the influence. The voltage of the secondary, due to this
influence, has a slight increase after a time period of
0,06 s and is permanently stable until the end of
simulation, figure 5.
Figure 6: It simulates the short circuit and only
controls the current of the secondary set to the nominal
current I "n  232A , for a real three-phase power
transformer SIMENS-ONAF (dimensions: length, width,
height 1140x800x1700 mm , transformer has fans which
start to work at higher loadings; oil flow depends only
on siphon-effect) with the following information:
connection Dy 5 , current of idle operation i0 %  1% ,
voltage of short circuit uk %  3,75% , voltages of
primary and secondary U np  U n1  10.5kV  ,
U ns  U n2  230V  , S n  160kVA , PFe  500W  ,
Pcun  3200W  , Vnp  Vn1  10.5kV  ,
Vns  Vn2  0.4kV .

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
10 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers

u % V12n 3.75 10.5 2  10 6

Xk  k   25.8
100 S n 100 160  103

The ratio of number of the coils of the primary and

secondary winding is:

U 1n 10,5
kt    45.652  46
U 2n 0.230

The reactance and dissipative inductance of primary

winding is:

 12.9
X k 25.8
X k1  
2 2

 0,043H 
X k1 12.9
L1  L 1  
 314

Magnetization impedance is:

V1nV1n V 21n
Zm  X m   
3  i0 % I nV1n i0 S n
2 6
 68,9k
10.5 10

0,01  160  10 3

Inductance of magnetization from the primary side is:

Fig. 5: Results of simulation no. 1: Time flow of flux, magnetization

68.9  103
 219,36H 
Xm
current and voltages of primary and secondary Lm  
 314
From the primary side the reactance of short circuit is:

Fig 6: Results of simulation no. 2: Short circuit is made by adjustment of loading in secondary Pcun  3200W  which was the goal of this
simulation

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers 11

Resistance of transformer’s primary winding: has been shown that such a computer simulation in a
qualitative way can describe the transient processes and
1 Pcun 1 Pcun 2 by doing so a better analysis of their behavior can be
R1  R p  R1k   V1n  performed. This paper described dynamic changes in
2 3I12n 2 S n2
real and ideal power transformer through MATLAB
10.5 210 6  6,89
1 3200 Simulink, which indicates possible malfunction or

2 160 210 6 defect.

Rbase 
Vn 2 
10.5 2  10 6
 689 References
Sn 160  10 3
[1] B. Mitraković: Transformatori, Beograd, 1985.

 2.194H 
Rbase 689 [2] Datasheet, online: March 10th 2011,
Lbase  
2f n 314 http://www.eps.et.tudelft.nl
[3] Van der Sluis, L., Rutgers, W.R., Koreman,
For this simulation a three-phase transformer ONAN C.G.A.: A physical arc model for the simulation of
was used, Dy 5 160 kVA  , 12(10.5)/0.4 kV  is single- current zero behaviour of high-voltage circuit
phase fed from the resource 50 Hz  . Transformer: breakers, IEEE T-PWRD 7(2), 1992, p. 1016–1022.
Nominal power 250e6,50 Hz  . Parameters of winding 1 [4] Akademik Antonin Veverka: TECHNIKA
(primary): 10.5e3 Vrms, R  0,01 p.u., L  0,02 p.u., VYSOKÝCH NAPĚTÍ SNTL/ALFA, DT621.3027
parameters of winding 2 (secondary): 0.4e3 3:621.315.31:II, Elektrickě razově jevy ve vinutih
Vrms/sqrt(3), active losses in the magnetic circuit: transforma torů a točivýh strojů, Praha, 1982.
50p.u. Characteristic of saturation: [0 0;0 1.0;1.0 1.22], [5] Mayer, D.: ŮVOD DO TEORIE ELEKTRICKÝH
residual flux = 0.6 p.u. OBVODŮ, SNTL/ALFA, 16. Nelinearni obvody
Input values which correspond to dialog box are: sa soustrednimi parametry, Praha, 1978.
[6] Kvasnica, J.: Teorie elektromagnetickeho pole,
6.89 Kapitola, V.: Kvayistacionárni elektromagneticke
R1 pu   0.01 p.u .
689 polé, ACADEMIA PRAHA, 1985.
[7] MATLAB SIMULINK Sim Power System,
0.043H  Copyright 1984-2002 The Math Works, Version
L1 pu   0,02 p.u .
2.194H  6.5.0,180913a, June 2, 2000.
[8] Bjelić, S., Marković, N. Jakšić, U.: The simplified
The short circuit was simulated with adjusted current procedure for calculation of influence of thermal
of secondary: losses on decrease of technical endurance of
electric equipment, III Energy efficiency in
160  103 application of electricity www.society-of thermal
 232A
Sn
I "n   engineering of Serbia, IEEP `2011, ISBN 978-86-
3Vn 3  0.4  103 7877-022-7, COBISS.SR-ID 184481804, 2011,
Abs. p. 28.
[9] Розенблат М.А.: Магнитные элементы
IV. Conclusion автоматики й вычислительной техники, УДК
538, Наука, Москва, 1974, p. 85–101.
The application of a computer simulation of dynamic
behavior of a power transformer in the software [10] Bjelic, S., Markovic N.: Multisliced-level type
program MATLAB-SIMULINK at transient processes, phase-comparison carrier relaying system for
the validity of a given theoretical model in extreme multi-terminal lines, 16th International
regimes as idle operation has been confirmed. Telekommunication forum, TELFOR 2008,
Belgrade, Session 7: AEE, work 29, Proceedings,
Measurement results a) in idle operation- time change 2008, p. 585.
of the flux verifies the value of expression (61) time
flow of a current of magnetization verifies the value of [11] Bjelic, S., Jaksic, U., Markovic, N.: Informatical
expression (59), and time flows of voltages of primary Model of the System of Automatic Breakdown
and secondary relations (47) and (48) at b) short circuit Control in Energetic Systems, ELEKTRONIKA IR
diagram verifies the value, figure 5. ELEKTROTECHNIKA, Vol. 1, 2011, p. 87–92.

In the first example of a transformer of high power [12] Markovic, N., Bjelic, S., Jaksic, U.: Development
and a second example of a transformer of low power it of new measuring systems based on symmetric
components in electric networks, Submission

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12
 12 Computer Simulation of Theoretical Model of Electromagnetic Transient Processes in Power Transformers

ID:37, Journal Arhive Kaunas University

Lithuania, ISSN 1392-1215,
ELECTRONICS&ELECTRICAL ENGINEERING,
T 120, No 8(104), 2010, p. 57–62.
[13] Marković, N., Bjelić, S., Jakšić, U.: Calculation of
influence of additional losses on decrease of time
period of exploitation of power transformer,
Inovacije i razvoj, Bor, br. 1, 2011, p. 75–86.
[14] Bjelić, S. et al: Estimation of currents flow, loss of
power and voltage fall down in distribution
network by using PDM KMp.q, 19 International
Conference of Electrical Distribution, Paper 0907,
Block 4.2, Method and tools, Cired, Wien, 2007.
[15] Bjelić, S.: Introduction in middle voltage networks
and low voltage installations, BOOK, ISBN 978-
86-83561-15-5, Cobiss SR-ID 14100 0204 SVEN,
Nis, 2007.

Authors’ Profiles
Slobodan Bjelić: Professor of
electrical engineering at Faculty of
Technical Sciences in Kosovska
Mitrovica, Serbia

Zorica Bogićević: Post-graduate

student for doctor degree in
electrical engineering at Faculty of
Technical Sciences in Kosovska
Mitrovica, Serbia

Copyright © 2014 MECS I.J. Information Technology and Computer Science, 2014, 01, 1-12

View publication stats