Proceedings of the 8th WSEAS International Conference on SYSTEM SCIENCE and SIMULATION in ENGINEERING
Modeling Ferroresonance in Single-Phase
Transformer Cores with Hysteresis
P. S. Moses, Student Member, IEEE, and M. A. S. Masoum, Senior Member, IEEE
paul.s.moses@gmail.com
Department of Electrical and Computer Engineering
Curtin University of Technology, Perth, WA, 6845, Australia
AbstractFerroresonance is a highly dynamic and nonlinear
power quality phenomenon caused by nonlinear inductances in
ferromagnetic materials and power system capacitances. It is
notorious for causing severe damage to power systems. This
paper carries out an investigation into single-phase transformer
ferroresonance initiated by switching transients. A hysteretic core
model is implemented which includes major and minor hysteresis
loops for the study of dynamic nonlinear phenomena. Time-
domain waveforms of transformer flux, voltage and magnetizing
current are computed from the model. Poincare and phase-plane
portraits are used to examine the stability domain of observed
ferroresonance modes.
Index TermsFerroresonance, hysteresis, nonlinear models,
power quality.
I. I NTRODUCTION
F ERRORESONANCE in power networks involving non-
linear transformers and capacitors has been well re-
searched for nearly a century. However, it is only in recent
years that nonlinear transformer modeling techniques have
begun to approach the level of sophistication required for
accurate ferroresonance studies. To that end, the importance of
hysteresis nonlinearities in dynamic and transient simulation
studies and its impact on the stability domain of ferroresonance
modes has recently been demonstrated [1][3].
Ferroresonance can be understood as a complex oscillatory
energy exchange between magnetic field energy of nonlinear
transformer/reactor cores and electric field energy of nearby
capacitances (e.g., series compensated lines or circuit breaker
grading capacitors). Without adequate dissipation through nor-
mal loads and losses, a substantial amount of energy sloshes
back and forth within a power system and manifests as over-
voltages and currents exhibiting high levels of distortion. This
has caused significant equipment damage in several cases and
continues to be a large safety hazard [4][6].
Over the years, research in ferroresonance has concentrated
into three main areas: (1) improving analytical methods and
transformer models, (2) development of transformer protection
and mitigation strategies and (3) case studies of system level
impacts [7], [8]. Despite the extensive literature available in
this area, ferroresonance continues to be a challenging problem
to analyze, predict and understand due to its highly nonlinear
and dynamic behavior. Researchers must adopt complex math-
ematical notions such as chaos theory to gain insight into this
phenomenon [9][11].
The four generally accepted ferroresonance modes which
can occur are (1) fundamental ferroresonance (period-1),
ISSN: 1790-2769
m.masoum@curtin.edu.au
(2) subharmomic ferroresonance (e.g., period-3), (3) quasi-
periodic ferroresonance and (4) chaotic ferroresonance. The
last two are non-periodic modes. There is also the possibility
of mixed modes or unstable modes where gradual system
variations or perturbations cause sudden jumps (known as
bifurcations) from one mode to another [12], [13]. From
a purely mathematical standpoint, these modes are due to
multiple competing solutions (known as attractors) to a sys-
tem of nonlinear differential equations of an electromagnetic
circuit. The nonlinearity is due to the magnetic properties of
ferromagnetic material.
In this paper, a single-phase transformer model including
hysteresis nonlinearity is implemented to study possible fer-
roresonance behavior (e.g., subharmonic modes). The model
is used to generate and plot time-domain waveforms including
flux, voltages and magnetizing currents, as well as Poincare
and phase-plane diagrams for selected values of series and
shunt capacitors initiating ferroresonance conditions.
II. H YSTERESIS C ORE M ODELING
The modeling of hysteresis has evolved significantly since
the 1970s when digital nonlinear core models were first being
developed. References [14], [15] provide a thorough historical
review of progress in hysteresis modeling. Early modeling
attempts indirectly incorporated hysteresis by the use of single-
value nonlinear inductors in parallel with a resistor represent-
ing eddy-current and hysteretic losses. Models evolved to use
families of ascending and descending curve functions for the
inclusion of major and minor loop effects of ferromagnetic
material. Some authors ignore minor loops and focus only on
major loops or make use of scaling factors on major hysteresis
loops to derive minor loops. These nonlinear approximations
are typically based on piece-wise, hyperbolic, trigonometric or
differential equations.
Today, there is still a tendency for transformer power quality
studies to ignore hysteresis and use anhysteretic approxima-
tions of the B H characteristic because of the modeling
complexity and computational burden associated with hystere-
sis nonlinearities. This simplification can be justified for some
studies because transformer design has improved over the
years and hysteresis loop widths have narrowed significantly to
appear almost anhysteretic. Therefore, for steady-state power
quality studies such as harmonic power flow in nonlinear trans-
formers, single-valued anhysteretic functions are considered
acceptable.
246 ISBN: 978-960-474-131-1
 Proceedings of the 8th WSEAS International Conference on SYSTEM SCIENCE and SIMULATION in ENGINEERING

On the other hand, for the study of dynamic, transient 5

and nonsinusoidal power system behavior, the representation Limiting
4 Descending
of minor hysteresis loop trajectories becomes important as
Branch (im )
additional operating points are created by the dynamic ex- 3

citation of a nonlinear hysteretic core. This is especially true

Core Flux Linkage (t) [V.s]

2 Limiting
in ferroresonance where major and minor loop trajectories can Ascending
1 Branch + (im )
potentially generate more ferroresonant operating points [3].
For this paper, a suitable nonlinear hysteretic core model of 0

a single-phase transformer is implemented to study ferroreso- 1

nance. 2

III. S INGLE -P HASE T RANSFORMER C ORE M ODEL 3

I NCLUDING H YSTERESIS N ONLINEARITY 4
A PSPICE computer model was developed for a single- 5
phase transformer with hysteresis nonlinearity (Fig. 2). A 1 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0.8 1
Magnetizing Current im (t) [A]
scalar hysteresis model based on [16] is implemented for
this work due to its simplicity in implementation for circuit
Fig. 1. The developed hysteresis model is shown under ramped sinusoidal
simulation. The magnetic flux density b and field intensity excitation to demonstrate the capability of the model to form major and minor
h are the main parameters for this model. However, for this hysteresis loops constrained by ascending (+ ) and descending ( ) limiting
paper, the hysteresis equations are modified such that flux hysteresis curve functions.
linkages and magnetizing currents are computed instead of
b and h (i.e., b and h im ). These parameters are more
accessible in PSPICE and easily measured from laboratory The fitting parameters and control the vertical and
tests. The modified hysteresis equations are as follows: horizontal scaling. The ascending limiting loop segment is
derived from (4) by shifting the function f (im ) to the right by
dim d 1 an increment of and similarly the descending loop function
=  
dt dt L + (im )(im ) d+ (im )
L0 is obtained by shifting the function to the left by (eqs. (5)-
0 (im )+ (im ) dim
(6), Fig. 1). Effectively, controls the width of the limiting
d
for 0 hysteresis loop.
dt
(1a) + (im ) = f (im ) (5)
dim d 1
=   (im ) = f (im + ) (6)
dt dt L + (im )+ (im ) d (im )
L0
0 (im )+ (im ) dim
The slope of the above ascending and descending functions
d
for <0 (1b) must be computed before the modified hysteresis equations (1)
dt
can be used. The ascending and descending slope functions
where + and are the limiting ascending and descending d (im )
are derived through differentiation of (5)-(6) with
dim
curve functions dependent on magnetizing current im . L0 respect to magnetizing current.
is the inductance or slope in the saturated region along the
limiting hysteresis curves. As noted from (1), the basis for d (im )
= (7)
this model is the use of slope functions for + and dim |im | + 1
to compute magnetization processes. This is derived from
Equation (1) is then computed from substitution of (2), (5), (6)
the assumption that domain wall motion density (Barkhausen
and (7). The next step is for the program to select one of the
jumps) are proportional to the growth of domain regions which
two equations (1) for didtm based on whether the magnetization
increase with field strength [16].
is increasing or decreasing. An IF statement in PSPICE is
The flux linkage and current relationships for the hysteresis
used to select one of the two equations in (1) based on the
model are
d (t) sign of the induced voltage (2). The magnetizing current im
e (t) = (2) is evaluated by integrating (1).
dt
There are two approaches to realize the above expressions in
i (t) = im (t) + ic (t) (3)
a PSPICE circuit. The first method is to implement a controlled
Before equation (1) can be computed, the ascending and voltage source for the induced voltage (2) based on (1). The
descending limiting hysteresis loop segments must be spec- second approach is more abstract but more numerically stable
ified. This paper proposes nonlinear analytical expressions to and is thus the favored approach. Equation (1) is realized by
describe the ascending (+ ) and descending ( ) limiting implementing a circuit loop consisting of an arbitrary capacitor
hysteresis loop segments. The proposed nonlinear function (4) with inherent current and voltage relationship together with a
can accurately be fitted to measured i characteristics of a controlled current source.
real nonlinear transformer.
dV d
f (im ) = sgn (im ) loge ( |im | + 1) (4) i=C =1 (8)
dt dt
ISSN: 1790-2769 247 ISBN: 978-960-474-131-1
 Proceedings of the 8th WSEAS International Conference on SYSTEM SCIENCE and SIMULATION in ENGINEERING

t=0.1s
Nonlinear Hysteresis 1
Core Model

v (t) (pu)
Rs Ls
e(t ) 0

i (t ) Cseries ic (t ) im (t ) 1
0 0.2 0.4 0.6 0.8 1
10
u (t )

i m (t) (pu)
Cshunt Rc
0

10
0 0.2 0.4 0.6 0.8 1
4

(t) (pu)
2
Fig. 2. Single-phase ferroresonance circuit with hysteresis core model 0
2
4
0 0.2 0.4 0.6 0.8 1
The capacitor voltage is in fact the flux linkage and its Time (sec)

current is governed by the controlled current source based on (a)

(1). A large resistance (1012 ) is placed in parallel with the
capacitor to suppress numerical convergence problems.
The developed hysteresis model can form major and minor
hysteresis loops based on specified + and functions. This
is demonstrated in Fig. 1 for linearly increasing sinusoidal
excitation.
IV. S IMULATION R ESULTS
The scenario under investigation is an unloaded single-phase
transformer operating under steady-state conditions interrupted
by a switch which is opened at t = 0.1s. This case is represen-
tative of single-phase fuse or circuit breaker action resulting
in a series ferroresonance circuit as shown in Fig 2. This can
also occur for three-phase transformer banks (i.e., 3 single-
phase transformers) where one of the phases has developed
a fault. The series capacitance could be grading capacitors in
circuit breakers, stray capacitances in bus-bars and feeders, or
possibly series reactive power compensation capacitor banks.
The impedance values and i hysteresis loop data of
the circuit are based on a real single-phase 440 V 50 Hz
transformer (Rs = 9.4 , Ls = 6.34 mH). The parameters
for the hysteresis model are listed in the appendix. In the
following simulations, selected values of Cseries and Cshunt
were chosen to generate different ferroresonance modes.
Fig. 3 demonstrates period-3 type ferroresonance initiated
by the opening of the switch at t = 1.0s when Cseries and
Cshunt are set to 10 and 38 F , respectively. The flux and
voltage waveforms show sustained harmonic distortions with
excessive magnetizing currents. The phase-plane and Poincare
diagrams indicate the system settling to a stable attracting limit
cycle.
For Cseries and Cshunt set to 10 and 22 F , harmonic
distortions have increased for flux and voltage waveforms (Fig.
4). Furthermore, the transient period from normal operation to
ferroresonance has a longer duration compared to the previous
case. The group of five dots in the Poincare map indicate this is
period-5 type ferroresonance. The phase-plane diagram shows
the presence of competing attractors in the stability domain.
Similarly, when Cseries and Cshunt are set to 29 and
38 F , respectively, the transformer exhibits extremely dis-
torted period-7 ferroresonance voltages (Fig. 5). The phase-
1.5
Poincare section
Time [sec] 1
0.5
0
10
5 2
0 1
0
d(t) 5
dt
1 (t) [pu]
10 2
(b)
Fig. 3. Subharmonic ferroresonance (Period-3) occurring at Cseries =
10 F and Cshunt = 38 F ; (a) time-domain waveforms indicating dis-
torted flux and voltages with large magnetizing current, (b) Poincare and
phase-plane diagrams indicating the system settling to stable steady-state
ferroresonance after switching transient.
plane portrait indicates the presence of multiple competing
attractors in the stability domain.
Finally, to emphasize the sensitivity of this phenomenon
to a small change in a system parameter, Cseries is varied
by a small amount from the previous case to 28 F and the
simulation is repeated (Fig. 6). The resulting waveforms show
unstable ferroresonance modes which appear to be period-
3 type lasting for only a few cycles before dampening out.
It is interesting to observe that due to the flux and voltage
relationship (2), the flux waveform takes a long time to
stabilize compared to the voltage waveform.
In some simulations, the developed PSPICE model had dif-
ficulties converging to a solution. This is most likely due to the
discontinuities presented by (1) at the point of magnetization
reversal when the model switches between equations. Prob-
lems can be circumvented somewhat by reducing the time-step
in the solver but at the expense of increasing processing time.

ISSN: 1790-2769 248 ISBN: 978-960-474-131-1

 Proceedings of the 8th WSEAS International Conference on SYSTEM SCIENCE and SIMULATION in ENGINEERING

1
v (t) (pu) 1

v (t) (pu)
0 0

1
1
0 0.5 1 1.5 2 0 0.5 1 1.5 2
10
i m (t) (pu)

i m (t) (pu)
0 0

5
10
0 0.5 1 1.5 2 0 0.5 1 1.5 2
4
(t) (pu)

(t) (pu)
2
0 0
2
2
4
0 0.5 1 1.5 2 0 0.5 1 1.5 2
Time (sec) Time (sec)

(a) (a)

1.5 1.5
Poincare section Poincare section

1 1
Time [sec]

Time [sec]

0.5 0.5

0 0
10 10
5 2 5 1.5
1 1
0 0 0.5
d(t) 0 0
5 d(t) 5 0.5
dt 1 (t) [pu] dt 1 (t) [pu]
10 2 10 1.5

(b)
Fig. 4. Subharmonic ferroresonance (Period-5) occurring at Cseries =
10 F and Cshunt = 22 F ; (a) time-domain waveforms for flux, magne-
tizing current and voltage, (b) Poincare and phase-plane diagrams showing
resulting steady-state ferroresonance with competing cyclical attractors.
(b)
Fig. 5. Subharmonic ferroresonance (Period-7) occurring at Cseries =
29 F and Cshunt = 38 F ; (a) time-domain waveforms indicating extreme
voltage and flux distortions, (b) Poincare and phase-plane diagrams showing
resulting steady-state ferroresonance with multiple competing attractors.

1
V. C ONCLUSION
v (t) (pu)

This paper demonstrates a single-phase transformer model 0

including hysteresis nonlinearity applied to the study of fer- 1

roresonance. The model is used to compute a number of 0 0.5 1 1.5 2

useful outputs such as phase-plane trajectories and bifurcation 10

im (t) (pu)

response, as well as, fluxes, voltages and current waveforms. It

0
is observed that the seemingly innocuous switching action can
lead to many different ferroresonance modes causing havoc in 10
0 0.5 1 1.5 2
a power system. A method for defining and modeling limiting
4
hysteresis loop segments for incorporation into an existing
(t) (pu)

2
hysteresis model exhibiting major and minor loop effects is 0
proposed and simulated in PSPICE. 2
4
0 0.5 1 1.5 2
A PPENDIX Time (sec)
H YSTERESIS M ODEL PARAMETERS
Fig. 6. Temporary ferroresonance condition for Cseries = 28 F and
Cshunt = 38 F . The waveforms exhibit Period-3 type ferroresonance for
= 0.7, = 1000, = 0.06, L0 = 0.07 the first few cycles after switch is opened before dampening out.

ISSN: 1790-2769 249 ISBN: 978-960-474-131-1

 Proceedings of the 8th WSEAS International Conference on SYSTEM SCIENCE and SIMULATION in ENGINEERING

R EFERENCES
[1] O. Bottauscio and M. Chiampi, Influence of hysteretic behaviour in
ferroresonant lcr circuits, in Proc. Digests of the IEEE International
Magnetics Conference INTERMAG Asia 2005, 2005, pp. 10791080.
[2] B. Patel, S. Das, C. Roy, and M. Roy, Simulation of ferroresonance
with hysteresis model of transformer at no-load measured in laboratory,
in Proc. IEEE Region 10 Conference TENCON 08, 2008, pp. 16.
[3] A. Rezaei-Zare, R. Iravani, and M. Sanaye-Pasand, Impacts of trans-
former core hysteresis formation on stability domain of ferroresonance
modes, IEEE Trans. Power Del., vol. 24, no. 1, pp. 177186, Jan. 2009.
[4] R. C. Dugan, Examples of ferroresonance in distribution, in Proc.
IEEE Power Engineering Society General Meeting, vol. 2, 1317 July
2003.
[5] D. A. N. Jacobson, Examples of ferroresonance in a high voltage power
system, in Proc. IEEE Power Engineering Society General Meeting,
vol. 2, 1317 July 2003.
[6] T.-P. Tsao and C.-C. Ning, Analysis of ferroresonant overvoltages at
maanshan nuclear power station in taiwan, IEEE Trans. Power Del.,
vol. 21, no. 2, pp. 10061012, 2006.
[7] E. F. Fuchs and M. A. S. Masoum, Power Quality of Electric Machines
and Power Systems. New York: Academic, 2008.
[8] M. Iravani, A. Chaudhary, W. Giesbrecht, I. Hassan, A. Keri, K. Lee,
J. Martinez, A. Morched, B. Mork, M. Parniani, A. Sharshar, D. Shir-
mohammadi, R. Walling, and D. Woodford, Modeling and analysis
guidelines for slow transients. iii. the study of ferroresonance, IEEE
Trans. Power Del., vol. 15, no. 1, pp. 255265, 2000.
[9] C. Kieny, Application of the bifurcation theory in studying and under-
standing the global behavior of a ferroresonant electric power circuit,
IEEE Trans. Power Del., vol. 6, no. 2, pp. 866872, 1991.
[10] B. Mork, Ferroresonance and chaos - observation and simulation of
ferroresonance in a five-legged core distribution transformer, Ph.D.
dissertation, North Dakota State University, 1992.
[11] B. Mork and D. Stuehm, Application of nonlinear dynamics and chaos
to ferroresonance in distribution systems, IEEE Trans. Power Del.,
vol. 9, no. 2, pp. 10091017, 1994.
[12] H. B. Thompson, J. M. T. Stewart, Nonlinear dynamics and chaos :
geometrical methods for engineers and scientists. Chichester West
Sussex: Wiley, 1986.
[13] J. Gleick, Chaos: Making a New Science. New York, NY: Viking,
1987.
[14] F. De Leon and A. Semlyen, A simple representation of dynamic
hysteresis losses in power transformers, IEEE Trans. Power Del.,
vol. 10, no. 1, pp. 315321, Jan. 1995.
[15] L. Chua and K. Stromsmoe, Lumped-circuit models for nonlinear
inductors exhibiting hysteresis loops, IEEE Transactions on Circuits
Theory, vol. 17, no. 4, pp. 564574, Nov 1970.
[16] J. Tellinen, A simple scalar model for magnetic hysteresis, IEEE Trans.
Magn., vol. 34, no. 4, pp. 22002206, 1998.

Paul S. Moses (S09) received his B.Eng. and B.Sc. degrees in Electrical
Engineering and Physics in 2006 from Curtin University of Technology, Perth,
Australia. He was awarded an Australian Postgraduate Award scholarship in
2009 and is presently working towards a PhD degree in Electrical Engineering.
His interests include nonlinear electromagnetic phenomena, power quality
and protection, renewable energy systems and power electronics. He has
also performed scientific research for the Defence Science and Technology
Organisation, Department of Defence, Australia.

Mohammad A.S. Masoum (SM05) received his B.S., M.S. and Ph.D.
degrees in Electrical and Computer Engineering in 1983, 1985, and 1991,
respectively, from the University of Colorado at Boulder, USA. Dr. Masoums
research interests include optimization, power quality and stability of power
systems/electric machines and distributed generation. Currently, he is an
Associate Professor at the Electrical and Computer Engineering Department,
Curtin University of Technology, Perth, Australia and a senior member of
IEEE.

ISSN: 1790-2769 250 ISBN: 978-960-474-131-1