ANALYSIS OF FERRORESONANCE

PHENOMENON IN TRANSFORMERS
Project report submitted in partial fulfillment of the requirement for the degree
of

Bachelor of Technology

Submitted by

1. GAURVI ARORA (1510110140)

2. RIA GUPTA (1510110299)

Under supervision of

Prof. Amit Bakshi

DEPARTMENT OF ELECTRONICS AND ELECTRICAL ENGINEERING

SCHOOL OF ENGINEERING
SHIV NADAR UNIVERSITY
(DECEMBER 2018)
 Candidate Declaration

We hereby declare that the thesis entitled “Analysis of Ferroresonance Phenomenon in

Transformers” submitted for the B Tech Degree program has been written in our words. We
have adequately cited and referenced the original sources.

(Signature)
GAURVI ARORA
1510110140
Date: __________

(Signature)
RIA GUPTA
1510110299
Date: __________

2
 CERTIFICATE

It is certified that the work contained in the project report titled “Analysis of Ferroresonance
Phenomenon in Transformers,” by “Gaurvi Arora and Ria Gupta,” has been carried out under
my supervision and that this work has not been submitted elsewhere for a degree.

Signature of Supervisor(s)
Prof Amit Bakshi
Electrical and Electronics Engineering
School of Engineering
Shiv Nadar University
December,2018

3
 ACKNOWLEDGEMENT

After dedicating all our hard work for this final year project, it has finally reached its
conclusion. First of all, we would like to express our gratitude to our project supervisor, Prof.
Amit Bakshi of SHIV NADAR UNIVERSITY, GREATER NOIDA for providing us with
guidance, help, and immense support in every little aspect, which has led us to the completion
of this project. We would also like to thank our final year project panel members Prof
M.P.Dave, Prof Dinkar Prasad, and Prof Himanshu S. Sekhar for their advice and guidance
to make this project successful. We are thankful to our families especially, our parents,
without whose love and support this couldn’t have been possible.

Last but not the least, we would like to express our appreciation to our friends for their
patience, understanding and their undivided support that they gave us throughout the
completion of our progress report and also to all those people who helped and involved
themselves in our final year project.

4
 TABLE OF CONTENTS

TITLE 1

CANDIDATE DECLARATION 2

CERTIFICATE 3

ACKNOWLEDGEMENT 4

TABLE OF CONTENTS 5

ABSTRACT 7

1. INTRODUCTION 8

1.1 Overview 8

1.2 Voltage Transformer (VT) 8

1.3 Power Transformer (PT) 8

1.4 General Criteria of Possible Ferroresonance in Power Systems 9

1.5 Ferroresonance in Three phase Systems 10

1.6 Scope of the Project 11

2. LITERATURE REVIEW 12

3. METHODOLOGY 13

4. REPORT ON THE PRESENT INVESTIGATION 14

4.1 Ferroresonance in Voltage Transformers 14

4.1.1 Transformer and System Configuration 14

5
 4.1.2 Non-Linear Dynamic Systems 16

4.1.3 Simulation Results 17

A) Case 1: Normal Sinusoidal Response 18

B) Case 2: Fundamental Frequency Ferroresonance 19

C) Case 3: Harmonic Ferroresonance 20

D) Case 4: Chaotic Ferroresonance 21

4.2 Ferroresonance in Power Transformers 22

4.2.1 Practical Example 23

A) Case 1: Periodic Solutions 26

B) Case2: Quasi-Periodic Solutions 27

C) Case 3: Chaotic Solutions 28

5. CONCLUSION 31

6. REFERENCES 32

7. APPENDIX (Base Values and MATLAB Codes for Paper I & II) 33

6
 ABSTRACT

The purpose of this project is to identify and analyze ferroresonance phenomenon in power
transformers and voltage transformers. Ferroresonance can be defined as a sudden
overvoltage and over current at a point of transient in the electrical power system in a short
period of time. Ferroresonance in voltage transformer can occur due to switching process or
lightning strike activities. Ferroresonance is difficult to analyze and the occurrence of the
phenomenon of ferroresonance is unpredictable.

Electrical equipment such as voltage transformer, capacitor voltage transformer and

switchgear that deals with ferroresonance phenomenon can cause serious damage to the
equipment. Analysis of Ferroresonance in the Transformer will be done using the software
MATLAB. The simulations will be recorded for the various cases with changing values of
the capacitor and resistance to find the condition at which ferroresonance occurs.

Throughout this project, ferroresonance activity will be analyzed and the condition at which it
occurs is considered.

7
1. INTRODUCTION
1.1 OVERVIEW
Ferroresonance phenomenon in voltage transformers can occur in a condition when a
nonlinear inductance that is the transformer cores is in series with the capacitance. The
capacitance can be the capacitance of transmission lines, cables, capacitance voltage
transformers or any shunt capacitors that is used in the power system. Other than that, is in a
condition when a lightning strike the voltage transformer. The high electric current that
carried by the lightning makes the voltage transformers receives a sudden high voltage that is
more than the rated voltage in the voltage transformers will make the voltage transformers
suffers a great external damage.

It is a phenomenon which can cause damage to electrical equipments of power systems by its
characteristic steady state over voltages and over currents which in excess of the normal
ratings of switching equipment.The need to analyse this phenomenon is due to its destructive
transient nature in power systems caused by the abnormal rates of harmonics or steady state
overvoltages and overcurrents in the electrical equipments.

1.2 Voltage Transformer (VT)

A voltage transformers (VT) or potential transformers (PT)is one of a type of instrument

transformers that are used in power system for the purpose of measurement of the voltage
output signals in medium or high voltage. It is used for stepping down the system voltage to a
safe value which can be fed to low ratings meters and relays. It also can be used as a
protective device’s and relay in a medium or high voltage network because of their low costs
and simple construction.

1.3 Power Transformer (PT)

Power transformers are used in distribution systems to interface step up and step down
voltages. It holds a low voltage, high current circuit at one side of the transformer and on the
other side of the transformer it holds high voltage low current circuit. The ratings of power
transformers range from low KVAs to very high MVAs.
8
1.4 General criteria of possible ferroresonance in power systems

Based on the general features of ferroresonance some basic criteria for occurrence of
ferroresonance can be stated:
• A capacitance in series with the core’s magnetizing inductance,
• unloaded or lightly loaded transformer (very rough rule of thumb: below 10-20% of the
rated load),
• single-pole switching procedures or interrupting devices in a three-phase system.

Transformers of any kind may be involved. Some obvious cases are:

• single-phase power transformers and voltage transformers energized through a series
capacitance,
• three-phase distribution transformers subject to single-pole switching or interruption,
• shunt reactors with saturable cores.

The involved capacitances can be in the form of actual capacitor banks, or as capacitive
coupling. Actual “capacitor banks” are easy to visualize, but capacitive coupling effects can
be more difficult to identify. Some examples where ferroresonance may occur are given
below.
• Series capacitor for line compensation.
• Shunt capacitor banks.
• Underground cables.
• Capacitive coupling, coupling capacitances between double circuit lines.
• Systems grounded only via stray capacitances.
• Grading capacitors on circuit breakers.
• Generator surge capacitors.
• Internal capacitive coupling in transformers.

9
1.5 Ferroresonance in three phase systems

Fig 1. Examples of ferroresonance in three phase systems

Ferroresonance in a three phase system may occur in the applied or induced voltage
connected to a capacitance in series with a transformer magnetizing reactance.

In Fig 1 three examples are given as examples when a single phase is switched in a power
system where wye-connected capacitance is paralleled with an unloaded wye-connected
transformer. The involved capacitors can be the shunt capacitances of the connecting cables
from the source or a shunted capacitor bank. The transformer phases are represented by jXm,

because the ferroresonance only involves the magnetizing reactance.

From Fig 1 it is seen that a series path through the capacitance(s) and magnetizing
reactance(s) is possible if one or two of the switches are open and if either the capacitor bank
or the transformer has grounded neutrals. If both neutrals are either grounded or not grounded
in Figure 5 no series path that includes at least one core and a capacitance exists, which then
10
means that ferroresonance is impossible. For the transformer core topologies in Figure 6
which provides a direct magnetic coupling between the phases, voltages can be induced in the
open phase(s) of the transformer. For the cases where ferroresonance is possible the above
holds for any configuration of the transformer core and even if each phase is supplied as
stand-alone units. In cases with a direct magnetic coupling between the phases ferroresonance
is possible even when there is no obvious series path from the applied voltage through a
capacitance and a magnetizing inductance

1.6 Scope of the Project

The project limits will cover simulation analysis of Ferroresonance in Transformer by using
the software MATLAB. The project was limited to software simulations as we were not able
to identify the given curve fitting values mentioned in the papers due to the unavailability of
the required lab equipment. Also due to the high value taken for the case of chaotic
ferroresonance in MATLAB, it cannot be replicated experimentally in the laboratory.
Thereby the result of the simulation will be analyzed to determine the characteristic of
ferroresonance phenomenon in the transformers.

11
2. LITERATURE REVIEW
In this chapter will cover a lot on the analyzing the characteristics of ferroresonance, and
ferroresonance in voltage and power transformers. The factors that can cause ferroresonance
in three phase transformer also will be identified. The modelling and simulation of the
voltage and power transformers are carried out by using one of the professional’s simulation
tools for analyzing power system transients that is MATLAB. There are many past studies
from the past journals in IEEE describing their studies for the transient analysis studies.
(ARAUJO, A.E.A , .MARTI, J.R., and SOUDACK, A.C.: ‘Ferroresonance in power systems:
chaotic behaviour’, IEE Proc. C, May 1993 and AMIN, Z. , .ZAHAWI, B.A.T.AL.,
AUCKLAND,D.W. and TONG, Y.K..: ‘Ferroresonance in electromagnetic voltage
transformers: A study based on nonlinear dynamics’, IEE Proc Gener Transm. Distrib,, July
1997). The analysis of the transients in transformers is basically modelled by a group of
important parameters to represent the elements that can contribute to the ferroresonance
phenomenon.

Ferroresonance is a general term applied to a wide variety of interactions between capacitors

and iron-core inductors that result in unusual voltages and/or currents. In linear circuits,
resonance occurs when the capacitive reactance equals the inductive reactance at the
frequency at which the circuit is driven. Iron-core inductors have a non-linear characteristic
and have a range of inductance values. In power system the ferroresonance occurs when a
nonlinear inductor is fed from a series capacitor. Ferroresonance phenomenon has been
studied using a 100 VA voltage transformer, fed through circuit breaker grading capacitance,
is studied using nonlinear dynamics methods, where the magnetising characteristic is
represented by a single-valued two-term polynomial of the order seven. The system exhibits
three types of ferroresonance: fundamental frequency ferroresonance, sub-harmonic
ferroresonance and chaotic ferroresonance. The paper shows that while fundamental
frequency and sub-harmonic ferroresonance can occur under practical operating conditions,
chaotic states are unlikely in reality. In a 25 MVA power transformer under consideration, the

12
solution of the nonlinear equation is exhibiting three types of ferroresonant states: periodic,
quasi-periodic and chaotic. Simulations of each of these states are presented for different sets
of values of capacitor and resistor. It is also shown that the transformer losses are an
important factor in obtaining the chaotic state.

3.Methodology
In this project we use the ode solver “ode45” to solve the differential equations.
MATLAB’s standard solver for ordinary differential equations (ODEs) is the function ode45.
Solve non-stiff differential equations. This function implements a Runge-Kutta method with a
variable time step for efficient computation.
The name ode45 comes from the fact that it is fourth order accurate globally and fifth order
locally.

Syntax : [T,Y] = ode45(odefun,tspan,y0)

 odefun :-
A function handle that evaluates the right side of the differential equations. All solvers solve
systems of equations in the form y′ = f(t,y) or problems that involve a mass matrix, M(t,y)y′ =
f(t,y).

 tspan :-
A vector specifying the interval of integration, [t0,tf]. The solver imposes the initial
conditions at tspan(1), and integrates from tspan(1) to tspan(end). For tspan vectors with two
elements [t0 tf], the solver returns the solution evaluated at every integration step. For tspan
vectors with more than two elements, the solver returns solutions evaluated at the given time
points. The time values must be in order, either increasing or decreasing.

 y0:- A vector of initial conditions.

The state variables used in the papers under consideration are z (1) and z (2), where
z (1) = λ

13
 z (2) = λ'

The reason we use ode45 than any other method is because of the fact that when the step size
is very small, then the results are satisfactory in case of ode45.
Due to the extra accuracy gain from the 4th order accuracy of Runge-Kutta (ode 45) far
outweighs the extra computational cost.

4. Report on the Present Investigation

4.1 Ferroresonance in Voltage Transformers

Ferroresonance occurs when one of the saturated core inductances matches with the
capacitance of the network system capacitance together with the nonlinear inductance of
power transformers combine to induce parasitic resonance which creates dangerously high
transformer voltages.

Recently, concern has been expressed about the possibility of ferroresonance in low capacity
electromagnetic voltage transformers (VTs) fed via capacitively graded circuit breakers.So in
this paper we consider the effect of capacitance of the power system circuit.That is how by
changing the shunt capacitance we get the condition for ferroresonance.

So the paper suggested applying chaos to the study of ferroresonance in electric power
circuits and covers the possibility of ferroresonance in power systems, particularly in the
presence of long capacitive lines.

In this paper, voltage transformer ferroresonance is analysed from the standpoint of nonlinear
dynamics. In particular, the effects of varying the value of system capacitance to earth on the
behaviour of the system are investigated.

4.1.1 Transformer and system configuration:

Figure 2 below shows the circuit diagram of system components at the 275kV substation
consisting of VT is a low thermal capacity wound voltage transformer. Ferroresonance

14
conditions occurred upon closure of disconnector DSI with CB and DS2 open, leading to a
system fault caused by failure of the voltage transformer primary winding.

Fig 2 System Diagram

where
CCBis circuit breaker CB grading capacitance
Cbbis the total busbar capacitance to earth

Fig 3 Reduced Equivalent circuit

Fig. 3 depicts the basic ferroresonance equivalent circuit used in this analysis.
Where
E is the RMS voltage applied to the circuit
ω is the supply frequency

15
λ is the voltage transformer flux linkage
Cseries is the circuit breaker grading capacitance
Cshunt is the total phase- to-earth capacitance including transformer windingcapacitance.
The resistor R represents transformer core losses shown in previous studies to be an
important factor in determining the behaviour of the system. Transformer current is
represented by the transformer nonlinear magnetisation characteristic, expressed in the form
of a power series. The paper implies the per unit values and the magnetisation curve was
modelled by a single-value seventh-order polynomial obtained by the least squares curve
fitting method from measured peak currents and voltages (λpeak = √2VRMS/ω).

i=aλ+bλ7 (1)

Where

a=3.42, b= 0.41 (in p.u. values) and λ is the transformer flux linkage in p.u. values.

The differential values for the above circuit is as follows:

1 dV 1 1
+ V+ (aλ + bλ7) =
ω dt Rω(Cseries+Cshunt) ω(Cseries+Cshunt)
Cseries
(Cseries+Cshunt)
√2 Ecosθ (2)

dλ dθ
=V =ω
dt dt

4.1.2 Non Linear dynamics system

The basic voltage transformer ferroresonance circuit was modeled by a differential equation
which determines the behavior of the circuit parameter (voltages, currents, magnetic flux
levels, as a function of time).Since the system is nonlinear in nature it explains the extreme
sensitivity of the system i.e. A small change in the due of system voltage, capacitance or

16
losses may lead to dramatic changes in the behaviour of the system. Furthermore the
nonlinear system may give rise to chaotic behaviour patterns.

4.1.3 Simulation Results

Here instead of using actual values of the circuit parameter E , ω , Cseries, Cshunt etc system
equations are made dimensionless by employing per unit values and the equation was written
as follows:

where g is the driving force amplitude and q is the damping parameter given by

1 dV 1 1
+ V+ (aλ + bλ7) =gcosθ (3)
ω dt q ω(Cseries+Cshunt)

Cseries
g=
(Cseries+Cshunt)
√2E (4)

q= Rω(Cseries + Cshunt) (5)

Values of E and ω were fixed at 1p.u., corresponding to AC supply voltage and frequency.

Cseries is the CB grading capacitance and its value depends on the type of circuit breaker
used.For the following analysis its value was fixed at a typical value of 0.5nf. Whereas the
value of Cshunt depends mainly on the length of busbar and the number of items connected to
it is considered to be around 10nF/metre.

In this analysis the value of Cshunt was varied over a range of realistic values for the site
under consideration, were assessed. to vary between 0.1nF and 5nF. Transformer core loss
was estimated between 50W and 500W. Calculations were performed for a value of
R = 225MΩ.

Simulation results for the four different parameter sets showing a normal sinusoidal response
plus the three possible types of ferroresonance conditions are presented below.

17
The initial conditions for the analysis was V(t)=√2 and λ(t)=0 at t=0representing circuit
breaker operation at maximum voltage which were calculated by substituting t=0 in the main
equation and simultaneously values of voltage was calculated at t=0.

CASE 1 : Normal sinusoidal response

Here the value of R=225MΩ.Cshunt =3nF and Cseries =0.5nF ; E and ω = 1p.u.
On putting the above value in (4) and (5) we get
q=247.4 and g =0.20
From the voltage graph obtained below we can infer the voltage value to be approximately
0.2 p.u

Fig 4 Transformer Voltage v/s Time Graph (Cshunt =3nF)

18
 Fig 5 Poincaré Map (Cshunt =3nF)

CASE 2 :Fundamental frequency ferroresonance

Here the value of R=225MΩ.Cshunt =1.25nf and Cseries =0.5nf
E and ω = 1p.u.
On putting the above value in equation 4 and 5 we get
q=123.7 and g =0.4
From the voltage graph obtained below we can infer the voltage value to be approximately 2
p.u.

Fig 6 Transformer Voltage v/s Time Graph (Cshunt =1.25nF)

19
 Fig 7 Poincaré Map (Cshunt =1.25nF)

CASE 3 :Subharmonic ferroresonance

Here the value of R=225MΩ.Cshunt =0.19nf and Cseries =0.5nf

E and ω = 1p.u.
On putting the above value in equation 4 and 5 we get
q=48.8 and g =1.02
From the voltage graph obtained below we can infer the voltage value to be more than 2 p.u.

Fig 8 Transformer Voltage v/s Time Graph (Cshunt =0.19nF)

20
 Fig 9 Poincaré Map (Cshunt =0.19nF)

CASE 4 :Chaotic ferroresonance

Here the value of R=1900MΩ.Cshunt =0.1nf and Cseries =3nf E and ω = 1p.u.
On putting the above value in equation 4 and 5 we get q=1850.4 and g =1.37
Also such high values of Cseries and R is unlikely but the waveform below demonstrates the
possibility in chaotic ferroresonance.
From the voltage graph obtained below we can infer the voltage value to be crosses 5 p.u. at
some time. The following is a case of chaotic ferroresonance.

Fig 10 Transformer Voltage v/s Time Graph (Cshunt =0.19nF)

21
 Fig 11 Poincaré Map (Cshunt =0.19nF)

4.2 Ferroresonance In Power Transformers

Ferroresonance is usually associated with potential (voltage) transformers that are used to
develop voltages used by relays; however, it can also occur with power transformers under
special circumstances. Although, such circumstances and transformer connections can be
simply avoided in practice, but if encountered can utterly destroy a transformer.

Ferroresonance is in principle a series resonance with a capacitor and a nonlinear inductance

that involves a saturable magnetizing inductance. The phenomenon preferably occurs in the
absence of an appropriate damping. In a power system that inductance typically consists of
the magnetizing inductances of power transformers and the series capacitances of occurring
capacitances and in the net.

The ferroresonance phenomenon can easily be demonstrated by a simple circuit comprising

an AC voltage source and an unloaded transformer connected in series with a capacitor.
Without the capacitor this circuit can look like in Figure 1 if one represents the transformer
by its equivalent circuit.

22
 Fig 12 An open transformer connected to an AC voltage source

The above circuit represents a 1.5 kVA dry-type 120/240 V transformer that is energized at
the 120 V winding. The current through the transformer comprises two superimposed parts.
The part that goes through the magnetizing inductance Lm is small except for when the core

goes in saturation and shows pronounced peaks near the voltage zero-crossings with a high
core flux. The other part goes through the resistor Rc and is responsible for the eddy current

and hysteresis losses. The winding resistances and winding leakage inductances are can be
assumed small, because their magnitudes are small relatively Lm and Rm. For an ideal

transformer Lm and Rm are infinite high. The resulting magnetizing current comprises high

order odd harmonics, but the peak current is comparatively small.

The situation, however, will differ considerably when a capacitor is placed between the
voltage source and the unloaded transformer as shown in Fig 12. The resulting exciting
current then will increase along with the induced voltage over the secondary which will be
also distorted. The reason for this behaviour is due resonance between Cs and Lm. In real

cases the exciting current relatively is smaller than in this example. Also it is questionable to
talk about resonance because the value the effective Lm varies because of its nonlinearity and

hysteretic behavior within each period such there will be no definite resonant frequency. The
resulting wave shapes, therefore, may jump from one or several quasi steady-states to
another.

23
 Fig 13 An open transformer connected to an AC voltage source via a capacitor C
s

4.2.1 Practical Example:

Ferro‐resonance configuration is formed when one or two of the source phases are lost while
the transformer is lightly loaded. This may be caused by single phase switching operations
such as clearing of single phase fusing and single phase reclosing.

In the following case, one phase of the system was open while other two are closed. This
leads to an induced voltage in the open phase because of capacitances in the system

Fig 14 Basic ferroresonant circuit

24
 Fig 15 Test System

Fig 16 Energized transformer phase

The system under study consists of source feeding an unloaded transformer with one of the
phase interrupted. Capacitive coupling with other two phases energizes the transformer.
Ferro‐resonance is due to interaction between the non‐linear magnetizing inductance of the
transformer’s open phase and capacitance of the distribution line.

The equivalent capacitance is found out by shorting first and second phase and omitting
transformer windings and ground capacitance of first and second phase. From figure 16, it is
observed that there will be no current on mutual capacitance between node 1 and node 2 since
they have the same potential therefore mutual capacitance between node 1 and node 2 can be
omitted as well. Mutual and ground capacitances which are connected to node 3 will be taken
for the equivalent circuit.
Hence, the equivalent capacitance is
C=Cg+2Cm (6)

ic=im+iR

25
 (7)
1 dφ
Where im=aφ+ bφ11 and iR= R dt

The magnetization characteristic of the transformer is simulated by the eleventh-order

polynomial, which depends upon the capacity of the transformer.
Here a=2.8 X 10-3 and b=7.2 X 10-3 are the fitting values of the saturation curve. First term
defines the linear region and the second one non-linear. φ refers to the flux in the transformer
core in p.u. value.

The differential equation for the basic ferroresonant circuit is:

d2 φ 1 dφ 1
+ + (aφ+ bφ11 ) = ωsEcos(ωst) (8)
dt2 RC dt C

ωs is the power frequency, 1.0 p.u.; E is the peak value of the voltage applied to the
equivalent circuit.

In the theory of dynamic systems, this equation is called a damped, forced nonlinear
oscillator. One of the most famous of these types of oscillators is the Duffing oscillator with a
cubic non-linearity, where the damping constant obeys and it is also known as a simple model
which yields chaos. This oscillator exhibits periodic solutions quasiperiodic solutions and
chaotic solutions, depending on the parameter values considered. Of the standard tools
available to study dynamic systems the most commonly used in the description of
quasiperiodic and chaotic is the Poincare map. The Poincaré map is simply a plot of the phase
plane at sampled intervals where the sampling frequency is the same as the driving
frequency. For purely periodic behaviour, then, the Poincaré map consists of a single point.
For systems with subharmonic components, the Poincaré map will consist of a finite number
of points.
This equation is solved through Ranga Kutta method to get solutions for various values of E,
R and C. Transformer has 25 MVA rated power and core loss given as %1 of rated
transformer capacity defining parallel resistance as 48.4 kΩ

CASE I : PERIODIC SOLUTIONS

26
 Fig 17 Transformer Voltage v/s Time Graph (C = 777 nF, R =48.4kΩ)

The transmission line taken here in this case is100 km long, corresponding to C = 777 nF,
R =48.4kΩ (Transformer losses =1%) and the equivalent source voltage is taken as
E = 0.15 p.u. Operation for this case occurs within the ferroresonant region of the
transformer.

Case II: QUASIPERIODIC OR PSEUDOPERIODIC SOLUTIONS

27
 Fig 18 Transformer Voltage v/s Time Graph (C = 1250.8nF, R = 1020 Ω)

To obtain this mode of behaviour, the transmission line’s length is taken to be 161 km long,
for C = 1250.8 nF; the transformer is considered virtually lossless (R = 1020 Ω) and the
equivalent source voltage was kept at E = 0.15 p.u.

Fig 19 Poincaré Map(C = 1250.8nF, R = 1020 Ω)

For a quasiperiodic system, the frequency components are not commensurate with the driving
frequency, and hence the Poincaré map evolves as a closed curve in the voltage flux plane.

Case III :CHAOTIC SOLUTIONS

28
Two sets of parameters were used for this case.
The first set is the same as for the first set of case (i) above, except that the transformer is
considered to be virtually lossless (R = 1020 Ω).
As in case I, the line length was 100km long (C=777nF) and E = 0.15 p.u.

Fig 20 Transformer Voltage v/s Time Graph (C = 777nF, R = 1020 Ω)

Poincare map, although having characteristics of a chaotic attractor, has an outline similar to
that of the quasi-periodic case. These results seem to illustrate the behaviour of a system on
the boundary between quasi-periodicity and chaos.

Fig 21 Poincaré Map(C = 1250.8nF, R = 1020 Ω)

29
The parameters for the second set consisted of R = 27 kΩ, which still represents a still
reasonable 1.6% transformer loss, a line length of 62 km (C = 483.44 nF), and E = 25.26 plu.,
which is a very high value of excitation voltage.

Fig 22 Transformer Voltage v/s Time Graph (C = 484.44 nF, R =27k Ω)

Fig 23 Poincaré Map(C = 1250.8nF, R = 1020 Ω)

In this case the Poincaré map evolves as a random set of points confined to a particular region
of the plane clearly indicating chaotic behaviour. For quasi-periodic systems, plots of the
dependent variables against time look at first glance, like non-periodic solutions.

30
However, their power spectral densities still indicate a finite number of spikes at non-
commensurate frequencies. In the chaotic case, the power spectral density will be broadband.

Although the simulation results of eqn. 2 indicate that all three types of behaviour, i.e.
periodic quasiperiodic and chaotic, can occur, chaotic solutions do not occur when realistic
values of transformer losses are considered. The same can be said for realistic values of the
source voltages, E. For realistic values of capacitance (line length) and source voltage, the
system described by eqn. 2 leaves the chaotic region at a transformer loss of 4.84 x 10-3 %
(R = 107Ω).
These results confirm the importance of not neglecting the transformer core losses in
ferroresonance studies. These losses, albeit small, can make the difference
between results showing chaotic behaviour and ordinary ferroresonance or linear operation.

31
5. CONCLUSION

• The system’s nonlinear differential equation has shown that a change in the value of the
equivalent circuit capacitance due to change in system configuration, can give rise to
different types of ferroresonance voltages.

• The paper describes how by reducing the value of the shunt capacitance ferroresonance
occurs and the value of voltage increases subsequently.

• Voltage transformers fed through a circuit breaker grading capacitance exhibits

fundamental frequency, subharmonic and chaotic ferroresonance conditions similar to high
capacity power transformers fed via capacitive coupling from neighbouring sources.

• We conclude the importance of not neglecting the transformer core losses in

ferroresonance studies. These losses, albeit small, can make the difference between results
showing chaotic behaviour and ordinary ferroresonance or linear operation(paper II).

• Chaotic as well as quasiperiodic ferroresonance has a broadband spectrum which could

cause serious problems of harmonic penetration in other locations of the power system.

• Also we can infer that while fundamental frequency and subharmonic ferroresonance
conditions may occur under common- place operating conditions, chaotic ferroresonance
states are not likely to occur under practical site conditions.

32
6. REFERENCES

1. AMIN, Z. , .ZAHAWI, B.A.T.AL., AUCKLAND,D.W. and TONG, Y.K..:

‘Ferroresonance in electromagnetic voltage transformers: A study based on nonlinear
dynamics’, IEE Proc Gener Transm. Distrib,, July 1997.

2. ARAUJO, A.E.A , .MARTI, J.R., and SOUDACK, A.C.: ‘Ferroresonance in power

systems: chaotic behaviour’, IEE Proc. C, May 1993.

3. EBRAY, ALI: “Parameter Study of Ferro-Resonance with Harmonic Balance Method”,

Royal Institute of Technology, Sweden, 2012

4.GREENWOOD, ALAN.: ‘Electrical Transients in Power Systems, 2nd Edition,

5. MOHAMMAD, A. S MASOUM, and .EWALD F. FUCHS..: ‘Power Quality in Power

Systems and Electrical Machines, 2nd Edition, 2008

6. MOZAFFARI, S , .HENSCHEL, S., and SOUDACK, A.C.: ‘Chaotic Ferroresonance in

power transformers’, IEE Proc Gen Transm Distrib, May 1995.

33
7. APPENDIX
Paper I
The base values used in the above analysis are base voltage = RMS phase voltage = (2751d3)kV
Base volt-amperes = transformer VA rating = l00VA
Base impedance = (base voltage)*/base volt-amperes = 252MQ
Base angular frequency = AC supply angular frequency= 2л50)rad/s
The base values used in the above analysis are
Base voltage = RMS phase voltage = (2751d3) kV
Base volt-amperes = transformer VA rating = l00VA
Base impedance = (base voltage)*/base volt-amperes = 252MQ
Base angular frequency = AC supply angular frequency= 2л50)rad/sV

Paper II
Base voltage : 63.5Kv
Base current : 131A
Base impedance : 484.73Ω

34