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

net/publication/236628205

Influence of the Settings of PSS2A and 2B Input Filters over the Damping of
Low-frequency Power Oscillations

Conference Paper · June 2011

CITATION READS

1 373

4 authors:

Yulian Rangelov Konstantin Gerasimov

Technical University of Varna ABB
79 PUBLICATIONS   58 CITATIONS    41 PUBLICATIONS   29 CITATIONS   

SEE PROFILE SEE PROFILE

Yoncho Kamenov Krum Gerasimov

Technical University of Varna Technical University of Varna
39 PUBLICATIONS   26 CITATIONS    59 PUBLICATIONS   39 CITATIONS   

SEE PROFILE SEE PROFILE

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

Modeling of Underground Transmission Systems View project

Intelligent energy and cost management systems (Originally in Bulgarian: Интелигентни системи за енергиен мениджмънт и управление на
разходите) View project

All content following this page was uploaded by Yulian Rangelov on 23 May 2014.

The user has requested enhancement of the downloaded file.

Proceedings of Papers

Volume 3

Serbia, Niš, June 29 - July 1, 2011

ICEST 2011 - XLVI INTERNATIONAL SCIENTIFIC CONFERENCE ON
INFORMATION, COMMUNICATION AND ENERGY SYSTEMS AND
TECHNOLOGIES, Serbia, Niš, June 29 - July 1, 2011

Proceedings of Papers - Volume 3 of 3 volumes

Editor: Prof. Dr. Bratislav D. Milovanoviü

Technical Editor: Dr. Zoran Ž. Stankoviü

Technical Co-Editor: Dr. Biljana P. Stošiü

Published by: Faculty of Electronic Engineering, University of Niš, Serbia

Printed by: UNIGRAF, Niš, Serbia

Number of copies printed: 280

Printing of this edition has been financially supported by Serbian Ministry of Science

ISBN: 978-86-6125-033-0
PO7.8 Design of Digital Control System of Spectrum Analyzer Built on MicroBlaze™ Processor ..............929
E. Stoimenov
Technical University of Sofia, Bulgaria
PO7.9 Verification of Improved Methodology for Design of Magnetic Components .....................................933
G. Nikolov, V. Valchev, E. Bekov, A. Marinov
Technical University of Varna, Bulgaria
PO7.10 Development of Industrial Circuits with Semiconductor Diodes and Optoelectronic Elements.......937
E. Koleva, I. Kolev, Ts. Karadzhov
Technical University of Gabrovo, Bulgaria
PO7.11 Curve Fitting for Sensors’ Analog Behavioural Modelling ....................................................................941
B. Nikolova, M. Todorov, T. Brusev
Technical University of Sofia, Bulgaria
PO7.12 Comparative Analysis of LCC Resonant DC-DC Converters.................................................................945
N. Bankov, G. Terziyski, A. Vuchev
Faculty of Electrical Engineering and Electronic, Plovdiv, Bulgaria
PO7.13 Overview of Automotive Network Protocols ...........................................................................................949
O. Stoyanov, G. Krastev, A. Stoyanov
University of Ruse, Bulgaria
PO7.14 A Matlab/Simulink Model of Piezoceramic Ring for Transducer Design..............................................952
I. Jovanoviü, D. Manþiü, V. Paunoviü, M. Radmanoviü, Z. Petrušiü
University of Niš, Serbia
PO7.15 Single-circuit and Double-circuit Regulating Apparatus for Gas Discharge Element........................956
S. Barudov, R. Dimitrova, M. Ivanova
Technical University of Varna, Bulgaria
PO7.16 Discharge Element with Transverse High-Frequency Excitation .........................................................959
S. Barudov, M. Ivanova
Technical University of Varna, Bulgaria
PO7.17 Investigation and Analysis of Organic Electroluminescent Heterostructures ....................................963
M. Aleksandrova ,G.H. Dobrikov , M. Rassovska,
Technical University of Sofia, Bulgaria

PO8 – POWER TRANSMISSION, DISTRIBUTION SYSTEM AND ELECTRICAL MACHINES

PO8.1 Possibilities to Manage Burning Process at the Conditions of Cement Kiln ......................................969
N. Krystev
Technical University of Sofia, Bulgaria
PO8.2 Damping Low-Frequency Oscillations by Three-Channel Power System Stabilizer PSS4B .............973
N. Nikolaev, Y. Rangelov, Konstantin Gerasimov
Technical Uniiversity of Varna, Bulgaria
PO8.3 Influence of the Settings of PSS2A and 2B Input Filters over the Damping of Low-Frequency
Power Oscillations .....................................................................................................................................977
Y. Rangelov, Konstantin Gerasimov, J. Kamenov, Krum Gerasimov
Technical University of Varna, Bulgaria
PO8.4 Modeling of Electromagnetic and Thermal Processes of High-Frequency Induction Heating
of Internal Cylindrical Surfaces of Ferromagnetic Details.....................................................................981
B. Aprahamian, M. Streblau
Technical University of Varna, Bulgaria
PO8.5 Model Research of Atmospheric Electric Effects in Electrical Low Voltage Network with
Local Photovoltaic System........................................................................................................................985
M. Vasileva, D. Dimitrov
Technical University of Varna, Bulgaria
PO8.6 Risk Assessment of Lightning Damages.................................................................................................987
M. Yordanova, M. Mehmed-Hamza, M. Vasileva
Technical University of Varna, Bulgaria

XVII
 Influence of the Settings of PSS2A and 2B Input Filters
over the Damping of Low-frequency Power Oscillations
Yulian Rangelov1, Konstantin Gerasimov2, Joncho Kamenov3, Krum Gerasimov4
Abstract – In the electric power system of Bulgaria power
system stabilizer type PSS2A and 2B are most common. They
were installed during the preparation period for connection with
the European electric power system. Usually the settings of these
stabilizers refer to the gain and time constants of the phase shift
block. This paper proves that the settings of the input filters can
influence the damping of local and mostly of inter-area
oscillations, which is a real problem after the actual connection
of the power systems. An algorithm for calculation of these
settings is proposed and results for single-machine and multi-
machine sample system are discussed.
 Ȧ is formed. Precisely this signal is entered input of the PSS
Keywords – Power system stabilizer, input filters, electric
power system.
I. INTRODUCTION
In order to highlight the influence of the input filters
(Fig. 1) of PSS-2A and 2B on the mechanical oscillations
damping, an algorithm for calculation of their settings is
developed.
The blocks with time constants from ɌW1 to ɌW4 form two
input filters (wash-outs) – one for the power channel and one
for the frequency channel. The purpose of these filters is the
prevent the PSS activation in cases of normal continuous
deviations of generator’s active power (P) and rotor speed
(Ȧ). The blocks with time constants T7 and T6 shift the signal
phase with lag up to 90°. The purpose of this is to obtain at
the end of the power channel a signal which is proportional to
the rotor speed deviation ¨Ȧ. The electric unbalance power
¨P which enters the channel can approximately be described
with the relation [2]:
d 'Z  'P
'P | TJ ˜ , ɬ.ɟ. 'Z |
s ˜ TJ
, (1)
from which for Ɍ7 | ɌJ the output signal of the block is
dt
proportional to minus ¨Ȧ. This signal is then added to the
incoming signal for Ȧ. The resultant signal enters in the filter
for torsional oscillation (ramp-tracking filter). For steam
1
Yulian Rangelov, Faculty of Electrical Engineering, Electric
Power Engineering department, Technical University of Varna,
Bulgaria, E-mail: j.rangelov@tu-varna.bg
2
Konstantin Gerasimov, Faculty of Electrical Engineering, Electric
Power Engineering department, Technical University of Varna,
Bulgaria, E-mail: kosio_gerasimov@abv.bg
3
Joncho Kamenov, Faculty of Electrical Engineering, Electric
Power Engineering department, Technical University of Varna,
Bulgaria, E-mail: kosio_gerasimov@abv.bg
4
Krum Gerasimov, Faculty of Electrical Engineering, Electric
Power Engineering department, Technical University of Varna,
Bulgaria, E-mail: k.gerasimov@tu-varna.bg
977
turbine generators, due to torsion forces in the shafts, rotor
oscillations occur in the sub harmonic range of few tens Hz.
Depending on the place of measurement of Ȧ deviations,
these oscillations superpose with the low-frequency machine’s
rotor oscillations (the rotor is treated as a single mass object).
If these oscillations are amplified and get into the excitation,
high-frequency oscillations will occur and as a result the rotor
can be damaged. The filter passes only inter-area oscillations.
The filter output is summed with the output of ¨P channel and
as a result a signal proportional to the inter-area deviation of
gain, i.e. the block with KS1. This way, regardless that there is
also ¨P channel, the PSS reacts to the inter-area deviation of
Ȧ.
The literature most often discusses only the filters settings
for passing a certain frequency range. Here, besides this
requirement, also is required that the filters output signal ¨Ȧɟ
is in phase the ¨Ȧ of the unit. This is a very important
requirement, especially when in PSS2A is entered only
generator bus voltage frequency deviation ¨f. If the output
signal ¨Ȧɟ is not in phase with ¨Ȧ, then a wrong phase
compensation will be obtained, even though the phase shifting
PSS block are tuned correctly.
II. ALGORITHM FOR CALCULATION OF
APPROPRIATE FILTER SETTINGS
In order to meet the requirement for close phases of ¨Ȧɟ
and ¨Ȧ, the following optimization task is formed.
J f (t ) ('Ze (ɉ, t )  'Z(t ))  g f (t );
min J f (t ), for J f (t ) d 0;
(2)
ɉ Ⱦ d ɉ d ɉȽ ,
ɉ
where ɉȾ and ɉȽ are respectively the lower and upper limits
of deviation the parameters to be ɉ={ɌW1, ɌW2, ɌW3, ɌW4,
Ɍ6, Ɍ7, Ɍ8, Ɍ9, ɄS2, ɄS3, ɄS4, M, N}; ¨Ȧɟ(ɉ,t) and ¨Ȧ(t) –
respectively the change in time of the filters output signal ¨Ȧɟ
and generators rotor speed ¨Ȧ caused by disturbances of the
AVR - ¨Uref; gf(t) – limit function which theoretically
converges to zero if there is complete phase and amplitude
overlapping of ¨Ȧɟ and ¨Ȧ.
To solve the optimization task (2), a computing scheme is
constructed in the MATLAB application Simulink, whose
structure is shown in Fig. 1. The solution is carried out by the
application Simulink Design Optimization (SDO).
 " Input  PSS 2 A "
Wash  outs
PSS

' U ref xc Ax  Bu 'P 'Ze 1  sTL1 1  sTL3 'VS

EPS

6
1  sTL4
sTW4
1  sTL2
sTW3 ()
1  sTW4
KS 2
1  sTW3 1  sT7
Cx  Du
KS 4 KS1
1
'P
( ) y ( ) to AVR

'f
2 90D LAG input
KS 3

('Z)
§ 1  sT8 ·
( )
6 ¨ ¸
'Z
N
sTW1
1  sTW1 © 1  sT9  ¹
sTW2 1
1  sTW2 1  sT6
()
¨ M ¸

R a m p  tr a c k in g

'Z
fi lte r
()

6
()
SDO
Fig. 1. Structure of the Simulink model used to calculation of PSS2Ⱥ filters settings

The EPS (Electric Power System) block represents the
model the electric power system in state space [1]. The
disturbances are introduced through change of the reference
¨Uref of the AVR in the generator whose PSS is being tuned.
The disturbance can be step or if the dominant modes of
mechanical oscillations for the specific generator are known
KS3=1,03 ɨ.ɟ., KS4=1,1 ɨ.ɟ., N=1,M=5. In Fig. 2 are shown
the frequency response of ¨Ȧɟ and ¨Ȧ, and in Fig. 3 – their
step response. It is clearly seen goal phase agreement of the
two signals is achieved.
0.08

0.07
the disturbance can be generated as a sum of sinusoids with 0.06

the frequencies of the dominant modes. At first the limits gf(t)

Magnitude (abs)

0.05

are set to about ± 0,005. In the process of the optimization 'Ze

'Z
0.04

procedure they are changed until the possibilities for their 0.03

satisfaction are exhausted. As a result the quantities of ɉ

which get 'Zɟ maximally close to 'Z are obtained.
0.02

0.01

0
270
The stated above is basis for the formulation of the
following algorithm for tuning of “Input – PSS 2A”:
'Ze
180
Phase (deg)

1. Construction of the computing scheme (Fig. 1) in

'Z
90

Simulink.
0
2. Use of SDO to solve the optimization task (2) for the
smallest limits gf(t). As a result the quantities of ɉ which get
'Zɟ as close as possible to 'Z are obtained.
-90

Fig. 2. Bode diagram of 'Zɟ /¨ Uref and ¨Ȧ/¨Uref

0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Hz
f

3. The calculated ɉ settings are used to generate the

frequency response characteristics of the input filters “Wash -
outs” and the torsional filter “Ramp - tracking”. Their 0.01

frequency bandwidths are checked if they meet the 'Ze

0.005

requirements. The input filters must not pass oscillations with

frequencies under 0,01 Hz, i.e. they must not pass through the 0

'Z
slow and continuous deviations in the regime parameters. The -0.005

torsional filter must not pass frequencies in the 5÷50 Hz -0.01

range, where the torsional oscillations are. If these

requirements are fulfilled, the calculated settings are treated as -0.015

final. Otherwise step 4. -0.02

4. The filters are tuned in respect to the frequency band -0.025

requirements. After that step 2 is carried out again but with

reduced number of parameters to be tuned ɉ={Ɍ6, Ɍ7, KS2, -0.03

Fig. 3. 'Zɟ and 'Z(t) for step change of ¨Uref with 2 %

0 2 4 6 8 10 12 14 16 18 20 s
t
KS3, KS4}. The obtained parameters settings are final.

III. TEST RESULTS

In Fig. 4, Fig. 5 and Fig. 6 are shown the frequency
responses of the input filters of inputs 1 and 2 and of the
For illustration of the proposed algorithm results are shown
torsional filter. It can clearly be seen the requirements for the
for the calculation of the settings of “Input-PSS 2A” of
pass-through frequency bands of the filters are satisfied.
generator G1 of the tested in [1] EPS. The calculated optimal
Therefore the calculated settings in step 2 of the algorithm are
settings are: TW1=3,66 s, TW2=3,66 s, TW3=2,07 s, TW4=0 s,
final.
Ɍ6=0 s, Ɍ7=2,07 s, Ɍ8=0,122 s, Ɍ9=0,193 s, KS2=0,16 ɨ.ɟ.,

978
 1
in Fig. 7. They, together with the step response of the regime
0.9 parameters U, P and Ȧ for step change with 1% of the AVR
0.8
reference (Fig. 8) shows how opening the filters for passing
through frequencies under 1 Hz enables the PSS to damp
0.7
them.
Hf
0.6
Magnitude (abs)

12
0.5

0.4 m Hf = 10.8163, fresonance = 0.10032

0.3
10

0.2

0.1

0 8
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Hz
f
Fig. 4. Frequency response of the input filter of PSS 2A input 1
1

V max
6
0.9

m Hf = 4.7853, fresonance = 0.10032

0.8

0.7

4
0.6
Magnitude (abs)

0.5

0.4
2
0.3

0.2

0.1
0
0 0.5 1 1.5 2
0 Frequency (Hz)
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Hz
f
Fig. 7. Transient processes quality generalized assessment
Fig. 5. Frequency response of the input filter of PSS 2A input 2 by the means of H’ norm
1

0.9 Step Responce

0.82
0.8

0.815
0.7
U [p.u.]

0.81
0.6
Magnitude (abs)

0.5
0.805

0.4 0.8
0 5 10 15 20 25 30 35 40 45 50

0.3 250

248
0.2
246
P [MW]

0.1
244

0 242
5 10 15 20 25 30 35 40 45 50 Hz
f 240

238
Fig. 6. Frequency response of the PSS 2A torsional filter 0 5 10 15 20 25 30 35 40 45 50

1.0003

In Fig. 7 and 8 are shown results revealing the difference in 1.0002

the transient processes quality for different values of the input 1.0001
w [p.u.]

filters parameters. With dashed line are depicted the results 1

for the settings: TW1=2 s, TW2=2 s, TW3=2 s, TW4=0 s, 0.9999

Ɍ6=0 s, Ɍ7=2 s, KS2=0,18 p.u., and with solid line – the 0.9998

settings: TW1=10 s, TW2=10 s, TW3=10 s, TW4=0 s, Ɍ6=0 s, 0.9997

0 5 10 15 20 25 30 35 40 45 50
Ɍ7=10 s, KS2=0,9 p.u. t [s] 
The generalized assessment of the transient processes
Fig. 8. Step response of the regime parameters U, P and Ȧ
quality [1,3], given by the H’ norms for both cases, are shown for step change with 1% of the AVR reference

979
 IV. CONCLUSION
The power system stabilizers type PSS2A and 2B
successfully damp the local low-frequency electro-mechanical
oscillations. The creation of large electric power systems
consolidations lead to higher requirement for these devices,
namely to damp successfully simultaneously the local and
inter-area oscillations. Because the latter and in the frequency
range under 1 Hz it is a must to tune appropriately the input
filters so that they can pass them through to the PSS.
REFERENCES
[1] Ƚɟɪɚɫɢɦɨɜ, Ʉ. Ɇɚɬɟɦɚɬɢɱɟɫɤɨ ɨɩɢɫɚɧɢɟ ɧɚ
ɟɥɟɤɬɪɨɦɟɯɚɧɢɱɧɢɬɟ ɞɜɢɠɟɧɢɹ ɜ ɟɥɟɤɬɪɨɟɧɟɪɝɢɣɧɚɬɚ
ɫɢɫɬɟɦɚ ɢ ɚɥɝɨɪɢɬɦɢ ɡɚ ɢɡɫɥɟɞɜɚɧɟ ɧɚ ɭɫɬɨɣɱɢɜɨɫɬɬɚ ɢɦ ɩɪɢ
980
View publication stats
ɦɚɥɤɢ ɫɦɭɳɟɧɢɹ. Ⱦɢɫɟɪɬɚɰɢɨɧɟɧ ɬɪɭɞ ɡɚ ɩɪɢɞɨɛɢɜɚɧɟ ɧɚ
ɧɚɭɱɧɚ ɫɬɟɩɟɧ ɞɨɤɬɨɪ ɧɚ ɧɚɭɤɢɬɟ. Ɍɍ-ȼɚɪɧɚ, 2006.
(Gerasimov, K. Mathematical modeling of electromechanical
transients in electric power systems and algorithms for analysis
of stability at small disturbances. Dissertation to acquire the
academic degree of Doctor of Science, TU-Varna, 2006)
[2] ɉɟɬɤɨɜ, ɉ., Ɇ, Ʉɨɧɫɬɚɧɬɢɧɨɜ. Ɋɨɛɚɫɬɧɢ ɫɢɫɬɟɦɢ ɡɚ
ɭɩɪɚɜɥɟɧɢɟ. Ⱥɧɚɥɢɡ ɢ ɫɢɧɬɟɡ ɫ MATLAB. Ⱥȼɋ Ɍɟɯɧɢɤɚ,
ɋɨɮɢɹ, 2002.
(Petkov, P., Konstantinov, M., Robust control systems. Analysis
and synthesis with MATLAB. ABC Tehnika, Sofia, 2002)
[3] “IEEE Tutorial Course Power System Stabilization Via
Excitation Control” Sponsored by IEEE Power Engineering
Society Life Long Learning Committee and the Excitation
Systems Subcommittee of the Energy Development and Power
Generation Committee
[4] “IEEE Recommended Practice for Excitation System Models
for Power System Stability Studies”, IEEE Standard 421.5 -
2005.