0% found this document useful (0 votes)
47 views

Student Member, B E E Senior Member, Ieee Suez Canal University Egypt Member, IEEE Fellow, Ieee

This document discusses methods for damping power system oscillations using global signals as inputs to power system stabilizers (PSS) and static var compensators (SVC). It proposes two global signals - the difference between generator speed deviations and tie line active power flow - to enhance damping of interarea modes. A two-level PSS controller is described that uses local signals for local modes and global signals selected by a coordinator for interarea modes. Alternatively, an SVC can be controlled using an appropriately chosen global signal superimposed over the voltage control loop to significantly contribute to system damping. The effectiveness of using global signals is tested on a 19-generator system similar to the Western North American grid.

Uploaded by

Rajesh Gangwar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
47 views

Student Member, B E E Senior Member, Ieee Suez Canal University Egypt Member, IEEE Fellow, Ieee

This document discusses methods for damping power system oscillations using global signals as inputs to power system stabilizers (PSS) and static var compensators (SVC). It proposes two global signals - the difference between generator speed deviations and tie line active power flow - to enhance damping of interarea modes. A two-level PSS controller is described that uses local signals for local modes and global signals selected by a coordinator for interarea modes. Alternatively, an SVC can be controlled using an appropriately chosen global signal superimposed over the voltage control loop to significantly contribute to system damping. The effectiveness of using global signals is tested on a 19-generator system similar to the Western North American grid.

Uploaded by

Rajesh Gangwar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

IEEE Transactions on Power Systems, Vol. 11, No.

2,May 1996

DAMPIN(; CONTROLLER DESIGN FOR POWER SYSTEM OSCILLATIONS


USIN(; G L O B A L SIGN A L S

Magdy E. Aboul-Ela A.A. Sallam James D. McCalley A.A. Fouad


Student Member, B E E Senior Member, IEEE Member, IEEE Fellow, IEEE
Suez canal University Iowa SLite University
Egypt US A

Abstract: This paper describes a new power system stabilizer are (1) difference between two generators speed deviation
(PSS) design for damping power system oscillations focusing (2) tie line active power flow. Although control
o n interarea modes The input to the PSS consists of two obtained remotely requires additional cominunic
signals The first signal is mainly to damp the local mode in the equipment, it is likely that the cost of such equipment WO
ruea where I'SS i\ located using the generator rotor speed as an offset by the additional operating flexibility gained h
input \ignaI The \econd is an additional global signal for control. In addition, we note that recent advances in the fi
damping interarea modes Two glohal signals are suggested, the phasor measurements may enable use of remote s
tie-lme active power and speed difference signals The choice of transmitted via satellite for control purposes [4].
PSS location, input signals and tuning 15 based on modal
analy\is and frequency response information. These two signals This paper extends investigation of using g
can also be used to enhance damping of interarea modes using signals reported in [5,6] via development of
SVC located i n the middle of the transmission circuit control strategies utilizing these signals The first \t
connecting the two oscillating groups. The effectiveness and utilizes a two-level PSS controller where the first lev
robustne5s of the new design are tested on a 19-generator control 15 derived from local signals and is desi
\ystem having characteristics and structure similar to the effectively deal with the local modes. The second
Western North American grid. control is supplied from a coordinator using selecte
states to deal with the poorly damped interearea modes
Keywords: Interarea owllation, PSS design, stability
The second strategy utilizes a static var compens
1 INTRODUCTION enhance damping of interarea modes Although S
often applied for the purpose of voltage support,
be used to increase damping of oscillations. We 1
The \tability ot electro-mechanical oscillations between that an SVC using local bus voltage inputs does n
interwnneLted synchronous generators is necessary for secure contiibute significantly to system damping
qy\tem opelation The oscillations of one or tnore generatois in significant contribution to system damping can
an area with repect to the rest of the system are called ZocuZ when an SVC is controlled by an appropriately ch
m o d e s , while those associated with groups of generators in signal superimposed over the voltage control loop
difteient rueas oscillating against each other are called interureu
triodes [ I ] Local modes are largely deteimined and influenced by 2 DAMPING OF SYSTEM MODES
local area states Inteiarea modes are more difficult tu study as
they iequire detailed representation of the entne interconnected
sy\tem and a e influenced by global state\ of larger areas of the
power network [2] This has led to an increased interere\t in the
nntuie of these mode, and m method\ (if contiolling them. N
i= AX+ Bu = Ax+ Blilj
J= 1
We are motivated to investigate more effective control rchemes
in order to incredse transmission capability for systems limited YJ = C/r
by oscillatory instability The most cominon control measures where B , and CJ are the colutnn-vector input matrix
in use today for this problem employ power system stabilizers
(PSS5) with local rotor speed as input. These PSSs are effective
in ctarnping local modes, and if carefully tuned [ 3 ] may also be
effective in damping interarea modes up to a certain
transini\sion loading. The effectiveness in damping interarea
modes is limited because interarea modes are not as highly
contiollable and observable in the generator's local signals as
the local modes.

We inve5tigate wing gZobuZ signals obtained remote from the


controller to damp interarea modes The global ugnals studied function residues [8]

2.1 Location of PSS Using PF


95 SM 520-7 PWRS A paper recommended and approved
by the IEEE Power System Engineering Committee of the
IEEE Power Engineering Society for presentation at
the 1995 IEEE/PES Summer Meeting, July 23-27, 1995,
Portland, OR. Manuscript submitted December 28, 1994;
made available for printing June 5, 1995.

0885-8950/96/$05.00 Q 1995 IEEE


768
Theretore, the machines with highest PF m the mode of interest
are good candidates for applying control via PSS. Therefore, the P
residue for the i-

2.2 Location of PSS Using Residues

The open-loop transfer function for a certain input/output of the


Ith generator of the system is

It can also he expiessed in terms of the modes and residues as

where R , i\the residue associated with ith mode and the jth
trander function (iJ It i s given by
selected generators,
conventional PSSs
Rzj = Slim
-tX, ( s - X % ) G ~ ( ~ ) (6) modes under stresse
is required particular
The residue I\ also given hy [ 6 ] . approach focusing o

capable of damping all


I? i j = Cif 1 V j Bj (7) next section.
the residue can be expressed in, terms of mode
y and observability
Equation (14) indicates that
Mode Controllahilitv The controllabihty of mode i from the j-
th generator is given by
maximum; therefore the
chosen as input to the
('OIZt Lj =I vi Bj I (8) must be directed towards
can be done by shaping
Mode Observahilitv. The tneasure of mode observability of a
certan niode 1 froin machine j ctan be defined as

ohsv,=/ c,tJ (9)

Fiom Equations (7),(8) and (9), it is clear that

1 I?, 1 =I qtl"&,I= cmt * obsvLI, (10)


AA, = AGi + j A W i =K Pssj
The residue associated with an eigenvalue AL and a feedback
transfei funLtion f/(s&,) , where K , 15 the constant gain of the Equation (16) shows
contio1ler, are ielnted by [SI of the i-th mode, due
to the real and imagin

Figuie 1 shows a PSS at the jth generatoi having a transfer


function H p , ~ . \ ~ ( s given
) by

H P S S J ( S ) = K PHS iS(~$ ) (12)

Then, replacing f;(sKI)in Equation (11) by HPSS/(S),


and f o r
small value\ of gain, it giveq

g the feedback to the system will cause a change in the For local modes, the
i-th eigenvalue Using Equations (12) & (13), this change is
given by
with the highest obse
the generator with the hi
AAl =I? H p s s J ( a L ) (14) the highert observabili
69
ccimhining information from different areas rather than being a
\ingle \tate of one generator. Therefore, iit will he called globul
signul, and it may be remotely transmitted. This signal will be
used as input to a two-level PSS controller, described in Section
4, and also to an SVC controller, described in Section 5. where,
T2j= -, T l j = ~T2j
j
4 PROPOSED TWO-LEVEL PSS DESIGN copq
The new PSS design aims to enhance the damping of poorly
damped local and interarea modes. This design provides a
control signal which is a sum of two component control
signals.
and m is the number of compensating blocks, T is a wash It
The firut control signal U’ is to provide damping for local
mode\ using the local generator rotor speed signal as a PSS time constant (usually 5-10 sec.), and 134
is the frequency If
input. The angle of compensation is computed at the local mode the local mode in rad/sec. The gam K I is taken as one-third e
frequency, and the controller time constants are chosen instability gain Kinst , i.e. K I = Kinst/-?.
accordingly. Thus, the local mode will be very highly damped.
This pwt of the controller is called PSSl and can he considered 4.2 PSSZ Design Using Tie-Line Power Signal
as a first-level controller i n a two-level control scheme as
shown in Figure 3. Each generator is considered as a subsystem. Interarea modes typically have high observability in the act e
The interaction between subsystems. represented in the power of tie-lines between areas involved in these intera a
interarea modes, is neglected in this stage. oscillations. Therefore, the tie-line active power can he used S
a stabilizing signal for damping the interarea modes associa d
The beconcl control signal is to provide damping for the with this tie-line. The selection of the controlled machine j : d
interarea modes, controlled from the selected machines using a the tie-line signal x is such that the transfer funct n
global input signal . The angle of compensation is computed at Ptie,/Vrefj has the largest residue magnitude for the intera a
the interarea mode frequency, with PSSl in service, and the mode of interest. From Equation (6), this is given by:
controller time constants are chosen to give this angle This
part of the controller is called PSS2. It is the second-level
controller or the coordinator as shown i n Figure 3 . The job of
thi\ controller is to obtiun the measurements characterizing the
global signals and send a control signal in terms of these
measurements to the selected machines for controlling the where j is the jth machine, and x is the tie-line number.
interarea modes. Therefore the total control signal for the jth
machine is The modal analysis is performed with all PSSls included, d
the residues associated with the lightly datnped interarea tno S
U, =U: + U; and several tie-lines are computed at the machines with le
highest controllability. These tie-lines are the heavily loat d
ties connecting the areas the generatois of which are swing g
against each other. The PSS2j transfer function is given by:
I Coordinator I

where T is the washout time constant, and Z l , 22 are le


compensating network time constants. The time constants .e
found using the same method described in Section 4.1. To stl Y
the effect of this controller on system modes using EPKI Sn 11
Signal Stability Program (SSSP), a user defined block )r
simulation of this controller is built [lo]. The gain Ki 1s
chosen to provide sufficient damping to the interarea mode If
concern.

From Equation (17) the total control signal (as shown in Fig .e
4) is given by:

Figure 3: Pioposed Two-Level PSS Design V sj =H p s s l j * A w j +H ps~2j*APtie (20)

4.1 Design of First-Level Controller, PSSl 4.3 PSSZ Design Using Speed Difference Signal

Since PSSl is intended for damping local modes of the The residue Rij, corresponding to the speed deviation signa >f
controlled machines, a PSSl is located at each machine with a generator j to control mode i, may not he large enougl- 0
poorly damped local mode. In this work, the amount of phase- achieve satisfactory dampmg. To increase the residue magnit le
lead required to shift the mode to the LHP is computed from either the observability or the controllability of these mode at
Equation (15) which is based on full system representation. The particular machines must be increased, according to Equa in
parameters of the lead/lag network are calculated frotn:
770

(10) Controllability can he increased by increasing the exciter controller IS given b


foiward gain, hut thic is not always good practice, as it can
reduce the \tability margm for the local mode.

We have observed that the right eigenvector elements


corre\ponding to the speed, and hence the observability vectors
of two machines J and k in generator groups oscillating against where 'Is ' '
2
s are the
each other (in an inteiarea mode), are always about 180 degrees
apart This suggests that the observability of this mode from
these two machines speed difference signal will be larger in
magnitude than each of the individual observabllities. If The transfer function between P
inteiaiea mode i is controlled from machine j using speed
difference signal of machines 1 and k then:

Ami
GEN i

10s (1+TIs) +
K 1 l G s (1+T 3,s)2

Figure4 New PSS Design


index for the i
AW,L = AWj - AWk (21)

The ohseivability index for this signal is given by

ohsv(dWjk)= c j V i - ckVi=(Cj-Ck)VI (22)

and the total contiol signal becomes


signal is given by:
Vs,=Hps,slj*AWj +Hpss2j*AWjk (23)
Vs = H svc AUjk
5 APPLICATION TO STATIC VAR
COMPENSATOR (SVC) Similarly, the transfer function -

The damping ohtaineii depends on SVC location in the system


and its control signal It has been found that the mid-point of
the tiansmisrion ciicuit, connecting the two oscillating groups
of- machines in the mterarea mode, IS the best SVC location for
damping enhancement [ 2 ] .An SVC can alter the P-6 curve by 4.1.
changing its output Bsv(7 (SVC suceptance) in such a way that
damping is increased. For example, for a one machine infinite
n, if the speed deviation of the machine is fed back as
iary signal to the SVC input, then Bsvc and In North power 'ys
subsequently the P-6 curve, will be altered such that the speed
deviation I \ d e c r c a d , and theiefore, damping can be increased.
This control stiategy has been called "Am-Control" ill]. We
u\e either tie-line active power signal or speed difference signal
as input to the SVC for dampmg interarea modes.

5.1 Tie-Line Power Signal to SVC

The SVC i s assumed to be located in the middle of a tie-line


connecting two areas oscillatmg against each other in a lightly
darnped interarea mode. This interarea mode will be highly
olhservable in such a tie line Therefore, the tie-line power of
ine is u r d as a stabihzing signal. A controller i s designed
der to compensate the phase lag of the system at the
frequency of interarea mode of interest. This SVC stabilizing
71
designed controller. This will he numerically shown for the test
system in Section 9.

7 ROBUSTNESS OF THE DESIGNED


CONTROLLERS

A rohust controller is expected to perform satisfactorily at


various operating conditions. HoweveI, since the propored
controller has been designed usmg linear analysis of the power
system, it i q important to check that it performs satisfactorily
when disturbances occur. The robustness of the controller is
evaluated in two different ways. The f m t is to evaluate the
system performance and damping ratio of critical modes of the
system under a heavily stressed operating condition compared
to the original operating condition. The second way is to
evaluate the system response to various disturbances using tune
domain simulation of the nonlinear power system model.

8 CASE STUDY
Figure 5: The T a t System
The single line diagram of the test system is shown in Figure 5.
This system has characteristics and structure similar to the control) to the selected machines having the largest resid1
North American Western grid (WSCC) [12]. It consists of 46 the damping is enhanced for all modes as shown in Tablt
bures and 19 machines in 7 areas Each area contains two or when using either the tie-line active power or the sp 1
three machines, one of which is represented hy the classical difference signals as input signals The control parameter.: f
model while the other one or two are represented hy a detailed PSS2s are given in Tahle 3. The controllers’ performance i
model (two-axis model with IEEE Type AC4 exciter, no robustness are tested by using the Extended Transient and b
governor, and a damping coefficient of 1 Opu). term Stability Program (ETMSP). From Figure 6, it is clear I t
system datnping is significantly improved.
Eigenanalyus, using the Small Signal Stability Program
(SSSP), of the open-loop system with no PSSs included To check controller robustness and then ability to stabilize
indicates there are 13 local and 5 intermea modes, all of which system, the system with all PSSls included has heen stres :d
are poorly damped (Table 1). It is seen that all machines are until it is unstable by increasing the transfer in the major 11 ZS
participating in local modes Therefore, a PSSl is located at and changing the load model to constant power. Figure 7 shc VS
each ot the 11 detailed machines It 15 noted that modes 12 & 17 the time simulation of this unstable point compared to the L se
helong to two equivalent machines and are not of concern. The when PSS2s are applied, using speed difference signal I 1s
control parameters of PSSls are computed and given in Table 2 clear that the addition of the PSS25 has stabillzed the sysl m
With F’SSls included (the conventional case) the 5 interarea under this unstable operating condition. Figure 8 illustrates le
modes are still hghtly damped By adding PSS2s (proposed expansion of the secure operating region for the two in; or
flows of the system due to the addition o f PSS2

Table 1: The Low Frequency Modes Of Oscillation With And Without Controllers
I PSSl I PSSl+PSS2 I PSSl+PSS2 I
772
Time R e s p o n s e o f R e a l p o w e r f l o w o f t i e l i n e 2 8 - 2 0
Tahle 2 The Fir\t-Level Controller\ Design (PSS 1) t o 0 . 0 1 s e c F a u l t a t Bus t7n
Local M ( d e Controllel F’arameter\ .__,

GEN (PSS1)

_ _- _ - PSSl
I 1
5 10 15 20 25 30 35 40

Time (Sec.)

Figuie 7. Stahilizatlon effect of PSS2

Secure Regions With PSSl Case and PSSI-PSSZCase

Time Response o f t i e - l i n e 2 8 - 2 0 r e a l power


Lo 0.01 s e c F a u l t a t B u s t 7 n

Northwest to Northern California flow, MW (Line 2627)


Figuic 8 : Expiuided opcratlng legion clue to PSS2

_______ NoConhol In addition, an SVC controller was designed, using tie-line


active power signal as input, to enhance damping of interarea
I-----
9 4 B I2 16 20 24 ZB 32 rnudes of the system. This controller was built in SSSP using
Time (Sec.) user tlefiried hlocks as this is a non standard controller.
Figure 6 : Time siniulation coinparison for tie-line 28-20 power
TwcI ptrtential locations are studied for the SVC :
Middle of line 28-20 (between Arizona and South
California areas) and signal from Ptie 28-20 (SVCI)
At B u d 2 2 (North California) and signal from F’tie 22-26
(SVC2)
Tahle 4 lists the residues for the interarea rnodes at SVCl and
SVC2. SVCl is chosen as it has much higher residues for mocles
#1,3,4,5. The effect of S V C l on darnping of these modes,
compared to PSS2 (of both types), is shown in Tahle 1. Figure
10 shows the hlock diagram of SVCl and its stahilizing
controller.
1.0. Q C/ Q L 10 - Q c / Q L

~ ~ ~ & - , B ~ v c ~ , u ~ PWRS, pp1076-1083, NOV. 1990.


1+ sT5
PaK-QL,[ 111 K.P.Padiyw, et al.," Damping Torque Analysis of Static

May 1991.

U,,,
or Kansas, Sept., 1994.
(4-
A9)
Figure 9: Block diagram of proposed SVCl controller BIOGRAPHIES

10 CONCLUSIONS

In this paper, a two-level control scheme for PSS design has


been proposed. The first level control is to provide damping for
the local modes using the local speed signal as an input to the
PSS. The second level is to enhance damping of interarea modes
using global signals as additional inputs to the PSS. The
proposed global signals are either the tie-line active power or
the speed difference signal. With proper choice of these signals
an increased residue magnitude for the interarea modes is
obtained, and hence these modes are effectively damped. The
proposed control scheme has been compared to the
the R.S. in Electrical Engineering (1950) at the Univers
Cairo; M.S. (1953) from the University of Iowa; and the
(1956) from Iowa State University. He is the 1990 t
Marston Distinguished Professor of Engineering at Iowa
: ;1
on
ate
conventional techniques using a 19Generator system. The University, and he is a Fellow of the IEEE.
results show that the proposed controller is more effective and James D. McCalley is Assistant Professor of Electrica ind
robust. These global signals have also been applied to SVC in Computer Engineering Department at Iowa State Univc 'ty,
older to damp the interarea oscillations, and the damping was where he has been employed since 1992. He worked for I. ific
significantly increased. Gas and Electric Company from 1986 to 1990. Dr. Mc( ley
received the R.S. (1982), M.S. (1986), and Ph.D ( 92)
10 REFERENCES degrees in Electrical Engineering from Georgia Tech. HI s a
registered professional engineer in California and a meml of
[ l ] M. Klein, G. J. Rogers, and P. Kiindur, "A Fundamental the IEEE.
Study of Inter-Area Oscillations in Power Systems", IEEE A.A. Sallam was born i n Mehalla El-Kobra, Egypt in 46.
Truns. PWRS, Vol. 6, No.3, Aug 1991, pp 914-921. He received the RS, MS, and Ph.D. from Cairo Univt ity,
[2] Yakout Mansour, "Application of ]Eigenvalue Analysis to Egypt in 1967, 1972, and 1976, respectively. He has ien
the Western North Americain Power System", employed with the Ministry of Industry, and the El-Nasr 1 ing
Eigenunulysis und Frequency Donmin Methods for System and Finishing Co., Department of Electrical Substa ns,
Dynumic P erfiormunce, IEEE 901'H0292-3PWR, pp 97- Mehalla El-Kohra, Egypt. In 1979, he jomed the Departm t of
104, 1989. Electrical Engineering, Suez Canal University, as a le^ rer .
[7] M. Klein, G. J. Kogers, S. Moorty, P. Kundur, "Analytical From 1981-1983 he was a visiting tnember of staff i the
Investigation of Factors Influencing Powei System Department of Electrical and Electronics Engine1 ng,
Stabilizers Performance", IEEE Truns. PWRS, Vol. EC-7, University of Newcastle upon Tyne, UK. Since 1983 h has
Sept 1992, pp 382-388. been workmg in the Depa tment of Electrical Engineering, uez
[ 4 ] Working Group H-7 of the Relaying Channel, Canal University, where he is now a Professor He is a uor
"Synchronized Sampling and P h a o r Measurements for tnember of the IEEE.
Relaying and Control", 93 WM 039-8 PWRS.
[SI X . Yang and A. Feliachi, "Statdization of Inter-area
Oscillation Modes through Excitation Systems," IEEE
Trans. on Power Systems, Vol. 9, No. 1, Feb. 1994, pp.
494-502.
[h] X. Yang, A. Feliachi, and K. Adapa,"Damping Enhancement
in the Wertern US Power System: A Case Study," SM94-
198
[7] A E. Hatnmad, "Analysis of Power System Stability
Enhancement by Static Var Compensators", IEEE Trans.-
PWRS ,Vol.l, No.4, Nov. 1986.
[8] F. L. Pagola, I. J. Perez-Arriaga , G. C. Verghese, "On
Sensitivities, Residues and participation. Application to
Oscillatory Stability Analysis and Control", IEEE Truns.
P W R S , V o l 4 , No 1, Feh.1989, p p 278-285
[9] D. R. Ostojic, "Stabilization of Multi-modal Electro-
mechanical Oscillation by Coordinated Application of
Power System Stabilizers", IEEE Truns. PWKS-6, pp
14'39-1445, 1991.
[IO] F'.Kundur, G.J. Rogers, D.Y.Wong , L. Wang, M.G. Lauhy,

You might also like