Computer Applications in Electrical Engineering

46

Susceptibility of electrical network to ferroresonance occurrence

Jzef Winiewski
Technical University of d
90-924 d, ul. Stefanowskiego 18/22, e-mail: jozef.wisniewski@p.lodz.pl

Edward Anderson, Janusz Karolak
Institute of Power Engineering
01-330 Warszawa, ul. Mory 8, e-mail: {edward.anderson,
janusz.karolak}@ien.com.pl


The calculations of electrical network susceptibility of chosen 110 kV, 220 kV and 400
kV networks to ferroresonance occurrence, its type and parameters were performed in the
study. The influence of such parameters as voltage transformer (VT) magnetization
characteristic, equivalent network capacitance, and the breaker grading capacitance on
ferroresonance was investigated. The calculations were performed using the EMTP/ATP
program. Some of the results were presented in this paper.

1. Introduction

The ferroresonance phenomenon in the power network has been known and described
for many years. However, it is difficult to investigate this phenomenon because of its
significant sensitivity to even small changes in network parameters [1, 2]. Also, the form
and parameters of the network equivalent scheme of devices like VT, power lines and
breakers exert an effect on the phenomenon character during computer simulations.
Looking for the range of network parameters, in relation to ferroresonance appearance by
means of the simulation method with gradually changing network parameters, is a long-
term process and does not guarantee finding a proper solution.
The field measurement method aiming at finding resistance of network to
ferroresonance is not effective because a slight change in investigation conditions
can significantly affect obtained results. Ferroresonance in the investigated network
can appear as a result of interaction between VT nonlinear inductance, network
capacitance, and a breaker grading capacitance [1, 2].
The simulation investigation of susceptibility of 110 kV, 220 kV and 400 kV
electrical networks to ferroresonance, its type, and parameters were carried out. The
calculations were performed using the EMTP/ATP program.
The influence of such parameters as VT magnetization characteristic, its burden,
equivalent network capacitance and the breaker grading capacitance on the
ferroresonance phenomenon was investigated. The opening of high voltage breaker
was the impulse which initiated the ferroresonance.
J. Winiewski, E. Anderson, J. Karolak, / Susceptibility of electrical network

47
2. Results of ferroresonance calculations

The scheme of the 110 kV network, taken under consideration for ferroresonance
calculation is shown in Fig. 1. After opening the breaker Q, the 110 kV network, i.e.
busbars and lines replaced by capacitance C
E
and inductive voltage transformer VT
is supplied by the grading capacitance C
Q
, which value depends on a breaker type
within the range of few hundred to few thousand pF.

VT
110 kV

110 kV
C
Q

CE
Q

Fig. 1. Scheme of the 110 kV network with ferroresonance phenomenon

The magnetization characteristic of 110 kV VT used in substation is based on
field measurements, Fig. 2.
0
200
400
600
800
0 0.05 0.1 0.15 0.2 0.25
i peak[A]
p
s
i

p
e
a
k
[
W
b
*
t
u
r
n
]

Fig. 2. The magnetization characteristic of 110 kV VT

The equivalent scheme of the investigated 110 kV network is presented in Fig. 3.
The scheme can be described by the set of equations:

+ =

= + + =
e e +
+
+ + + = =
+

) 1
C
C
( R K ;
R C
1
K ); 1
R
R
( ) 1
C
C
( K
)]; t cos( U
C
) ( i
)] a n a ( K K [ U [ K / 1
dt
dU
; U
dt
d
Q
E
s 3
p Q
2
p
s
Q
E
1
m
Q
2 1 n
n 1 3 2 1
(1)
where: + -flux linkage, ) t cos(
m
U ) t ( u e = - source voltage,
3 / 2 * kV 110
m
U = ; O = k 3 . 30
s
R ; O =
9
10
p
R .
J. Winiewski, E. Anderson, J. Karolak, / Susceptibility of electrical network

48

u(t)
Rs
Rp
C
Q

+(i
2
)
i
i
2
i
1
i
3
C
E
VT
Q

Fig. 3. Equivalent scheme of the investigated 110 kV network taken for ferroresonance calculation

Magnetization characteristic of VT can be described by the equation:
n
n
a
1
a ) (
2
i + + + = + (2)
where:
6
10 17 . 3
1
a

= ;
16
10 025 . 1
n
a

= ; 5 n = .
The graphical solution of the equations describing the network for steady state
for chosen parameters is presented in Fig. 4. The intersection of the u vs. i network
characteristic and the U
m
line representing the source voltage determines the possible
operation points 1, 2 and 3.

-600
-400
-200
0
200
400
600
-200 -150 -100 -50 0 50 100 150 200
i[mA]
u[kV]
u
m

1
2
3

Fig. 4. Possible operation points. Solution for CQ=1000pF, CE=2000pF
(1 and 2 - stable operation points, 3 - unstable operation point)

The graphical method, though simple, has a numerous limitations [1, 2, 3]. More
precise results, more similar to those observed in the network, provide the
J. Winiewski, E. Anderson, J. Karolak, / Susceptibility of electrical network

49
calculations by using the EMTP program. For different sets of parameters, the
network behavior after the breaker opening was performed.
In the investigated range of C
Q
and C
E
capacitance variation, ferroresonance did
not appear or it appeared as transient short-time or stable with network frequency of
f
s
= 50 Hz, 1/3f
s
, 3f
s
or stable with chaotic shape.
For example, Fig. 5 shows the transients of phase voltages, VT currents, residual
voltage and Fourier transform of phase voltages for ferroresonance with frequency
of 1/3f
s
and Fig. 6 presents the ferroresonance with a chaotic shape.


0E+0
2E+4
4E+4
6E+4
8E+4
0
5
0
1
0
0
1
5
0
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
4
5
0
5
0
0
(f ile juk123_3f az.pl4; x-v ar t) v :DA v :DB v :DC
0.0 0.2 0.4 0.6 0.8 1.0 1.2 [s]
-90
-45
0
45
90
[kV]
(f ile juk123_3f az.pl4; x-v ar t) c:DA -EA c:DB -EB c:DC -EC
0.0 0.2 0.4 0.6 0.8 1.0 1.2 [s]
-70
-35
0
35
70
[mA]
(f ile juk123_3f az.pl4; x-v ar t) v :3U0_1
0.0 0.2 0.4 0.6 0.8 1.0 1.2 [s]
-40
-20
0
20
40
[V]
Phase voltages
on VT
connectors
Phase currents
in VT's
Residual
voltage 3U
0
on
VT connectors
Fourier transform of
phase voltage on VT
connectors
u [kV]
f [Hz]

Fig. 5. Phase voltages, VT currents, residual voltage (3U0) and Fourier transform of phase voltage
on the secondary side of VT during the stable ferroresonance with frequency of 16.6 Hz
(CQ=300 pF, CE= 250 pF)
J. Winiewski, E. Anderson, J. Karolak, / Susceptibility of electrical network

50

0E+0
2E+4
4E+4
6E+4
8E+4
0
5
0
1
0
0
1
5
0
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
4
5
0
5
0
0
u [kV]
f [Hz]
0.0 0.2 0.4 0.6 0.8 1.0 1.2 [s]
-200
-100
0
100
200
[kV]
0.0 0.2 0.4 0.6 0.8 1.0 1.2 [s]
-0.70
-0.35
0.00
0.35
0.70
[A]
0.0 0.2 0.4 0.6 0.8 1.0 1.2 [s]
-150
-75
0
75
150
[V]
Phase voltages
on VT
connectors
Phase currents
in VT's
Residual
voltage 3U
0
on
VT connectors
Fourier transform of
phase voltage on VT
connectors

Fig. 6. Phase voltages, VT currents, residual voltage (3U0) and Fourier transform of phase voltage
on the secondary side of VT during the stable chaotic ferroresonance (CQ=2000 pF, CE= 1000 pF)

The results for the complete range of investigated parameters C
Q
and C
E
were
placed on the ferroresonance map, whose fragment is presented in Fig. 7. It shows
the areas free of ferroresonance (N), with stable chaotic ferroresonance (T/Ch), with
stable harmonic ferroresonance (e.g. T/16 Hz), or ferroresonance disappearing after
a short time (e.g. Z/0.9s). The maximal observed values of ferroresonance
overvoltages are also visible. The map is calculated for unloaded voltage
transformers.
The similar ferroresonance maps, but for voltage transformers loaded with
resistance 10 O (in the open triangle of secondary coils) are presented in Fig. 8. The
map of ferroresonance for the 400 kV network with unloaded voltage transformers is
shown in Fig. 9.
J. Winiewski, E. Anderson, J. Karolak, / Susceptibility of electrical network

51

T/16,6Hz
70
T/16,6Hz
80
T/Ch
100
T/Ch
120
Z/1
45
T/Ch
120
T/Ch
130
T/Ch
140
T/Ch
150
T/Ch
160
T/Ch
160
T/Ch
160
T/Ch
170
T/14Hz
72
T/Ch
100
Z/1
110
T/10Hz
40
T/Ch
135
T/Ch
150
T/16,6Hz
160
T/Ch
160
T/Ch
180
T/Ch
180
T/Ch
180
T/Ch
200
T/16,6Hz
80
Z/0,6
99
N
T/Ch
140
N T/Ch
140
T/Ch
160
T/Ch
170
T/Ch
170
T/Ch
200
T/Ch
180
T/Ch
200
T/Ch
200
T/16,6Hz
80
N N
T/Ch
140
N T/Ch
170
T/Ch
170
T/Ch
200
T/Ch
260
T/Ch
270
T/Ch
270
T/Ch
270
T/Ch
270
T/16,6Hz
100
T/Ch
100
N
T/Ch
150
N N
T/50Hz
250
T/50Hz
250
T/50Hz
250
T/50Hz
260
T/50Hz
260
T/50Hz
270
T/50Hz
270
N N Z/0,5
150
T/50Hz
225
T/16,6Hz
72
T/50Hz
235
T/50Hz
230
T/50Hz
250
T/50Hz
255
T/50Hz
260
N T/50Hz
275
T/50Hz
280
Z/0,9
80
N T/50Hz
215
T/50Hz
200
T/16,6Hz
70
N
T/50Hz
240
T/50Hz
250
T/50Hz
255
N
T/50Hz
265
T/50Hz
275
T/50Hz
280
CE [pF]
CQ [pF]
250
500
1000
1500
2000
3000
1000 300
2500
600
T/Ch
130
2000 3000 4000 5000 6000 2500 3500 4500 5500 1500

Fig. 7. The ferroresonance map for the investigated 110 kV network for unloaded VT
(values of overvoltage in [kV])

T/16,6Hz
70
N Z/0,5
100
Z/0,8
110
N Z/0,4
120
T/Ch
140
Z/0,5
130
T/Ch
140
T/16,6Hz
130
T/Ch
140
T/Ch
170
T/Ch
160
T/16,6Hz
70
Z/0,3
90
Z/0,4
105
N T/Ch
125
T/Ch
150
T/16,6Hz
120
T/16,6Hz
120
T/Ch
165
T/Ch
180
Z/0,5
170
Z/0,5
195
T/16,6Hz
80
Z/0,4
100
T/Ch
110
T/Ch
130
N Z/0,2
140
Z/0,2
160
T/Ch
160
T/Ch
170
Z/0,3
160
Z/0,7
185
T/Ch
200
Z/0,9
210
T/16,6Hz
75
T/16,6Hz
90
Z/0,2
110
Z/0,9
150
N T/Ch
145
Z/1
170
T/Ch
190
T/Ch
200
T/Ch
200
T/Ch
210
T/50
280
T/Ch
310
N Z/0,2
110
T/Ch
125
Z/0,5
160
N Z/0,4
190
Z/0,3
200
Z/0,3
190
T/50Hz
260
T/50Hz
265
T/50Hz
270
T/50Hz
270
T/50Hz
280
Z/0,4
100
N Z/0,6
145
T/50Hz
230
N T/50Hz
230
T/50Hz
245
T/50Hz
250
T/50Hz
255
T/50Hz
260
T/50Hz
265
T/50Hz
275
T/50Hz
280
Z/0,4
100
Z/0,3
120
Z/0,3
180
T/50Hz
220
N T/50Hz
235
T/50Hz
240
T/50Hz
250
T/50Hz
250
T/50Hz
260
T/50Hz
270
T/50Hz
270
T/50Hz
280
CE [pF]
CQ [pF]
250
500
1000
1500
2000
3000
1000 300
2500
600
Z/0,5
120
2000 3000 4000 5000 6000 2500 3500 4500 5500 1500

Fig. 8. The ferroresonance map for the investigated 110 kV network for voltage transformers loaded
with resistance 10 O (in opened triangle of secondary coils)
J. Winiewski, E. Anderson, J. Karolak, / Susceptibility of electrical network

52
CE [pF]
CQ [pF]
250
500
1000
1500
2000
3000
2500
1000 2000 3000 4000 5000 6000 600 1500 2500 3500 4500 5500 300
Z/0,2
220
T/16,6
280
T/16,6
310

T/16,6
360

N

T/16,6
390

T/16,6
420

T/16,6
440

T/16,6
470

T/16,6
490

T/Ch
390

Z/0,4
380
Z/0,4
360
T/16,6
240

T/16,6
330

T/16,6
330

T/16,6
370

N T/16,6
400

T/16,6
430

T/16,6
450

T/16,6
480

T/16,6
500

T/Ch
390
Z/1,2
380
T/50
330
T/Ch
290
T/Ch
280
Z/0,2
370

T/16,6
450

N T/16,6
420

T/16,6
440

T/16,6
470

T/16,6
490

T/16,6
510

Z/0,5
380

Z/0,2
360
T/50
340
Z/0,3
290
Z/0,1
350
N T/Ch
500
N T/Ch
540
T/Ch
540

T/16,6
490

T/16,6
500

Z/0,2
380

Z/0,2
360

Z/0,6
360

T/50
360
Z/0,3
360
N T/Ch
540
T/Ch
650
N T/Ch
610
T/Ch
620
Z/0,2
660
Z/0,3
370
T/16,6
570

T/16,6
560

Z/0,4
310

Z/0,2
310

Z/0,2
460
Z/0,4
540
Z/0,2
490
Z/0,2
720
N T/Ch
760
T/25
800
T/Ch
700
Z/0,2
320
Z/0,2
320
Z/0,2
310
Z/0,2
310
T/50
320
Z/0,3
480
Z/0,7
760
T/Ch
900
T/Ch
900
Z/0,3
310
Z/0,3
350
T/Ch
820
Z/0,2
320
Z/0,2
320
Z/0,2
320
Z/0,5
330
T/50
330
Z/0,2
320

Fig. 9. The ferroresonance map for the investigated 400 kV network for unloaded VT
(values of overvoltage in [kV])

3. Conclusions

The results of the simulation calculations of the investigated phenomenon, its
properties and parameters for 110 kV substation with unloaded VT were presented
as ferroresonance maps. Similar calculations for 220 kV and 400 kV networks with
different VT types were performed. The VT parameters were measured in the testing
station or were obtained from producers. The calculated results were partly verified
by comparing them to those of network measurements. The study results can be
useful for planning substation work and also for explaining VT or other network
devices faults occurring in the past caused by ferroresonance overvoltages.

References

[1] Ferraci P., Ferroresonance. Cahier technique. No. 190, Groupe Schneider, 1998.
[2] Jacobson D.: Field Testing, Modelling and Analysis of Ferroresonance in a High Voltage
Power System. Ph.D. Thesis, University of Manitoba, Winnipeg, Canada 2000.
[3] Valverde V., Mazon A.J., Zamora I., Buigues G.: Ferroresonance in Voltage
Transformers: Analysis and Simulations. International Conference on Renewable
Energies and Power Quality (ICREPQ'07) Sevilla, Spain.
[4] Wisniewski J., Anderson E., Karolak J.: Search for Network Parameters Preventing
Ferroresonance Occurrence. 9th International Conference Electrical Power Quality
and Utilisation, 2007, Barcelona, Spain.