The Impact of Closed-loop Power Flow Control Strategies on Power System Stability Characteristics in a Single-generator System-2006

7/25/2019 The Impact of Closed-loop Power Flow Control Strategies on Power System Stability Characteristics in a Single-ge

1/9

34 SOUTH AFRICAN INSTITUTE OF ELECTRICAL ENGINEERS Vol.97(1) March 2006

THE IMPACT OF CLOSED-LOOP POWER FLOW CONTROL

STRATEGIES ON POWER SYSTEM STABILITY CHARACTERISTICS IN A

SINGLE-GENERATOR SYSTEM

A. Ally and B. S. Rigby

School of Electrical, Electronic and Computer Engineering, University of KwaZulu-Natal, Durban 4041

Abstract: This paper presents a theoretical study into the influence of closed-loop control of ac power flow on

the small-signal and transient stability characteristics of a single-generator study system. Both the constant power

and constant angle modes of power flow control are examined for a range of controller response times. The results

indicate that the effect of a power flow controller on system stability is dependent on both the mode of the controller

and its response time.

Key words: Power Flow Control, Small-Signal Stability, Transient Stability, Thyristor Controlled Series Capacitor.

1. INTRODUCTION

The broad objective of Flexible AC Transmission Systems

(FACTS) is to enhance the controllability and power transfer

capability of interconnected ac power systems by means of

power electronically controlled compensators [1]. Within

this broad objective, FACTS devices can be used in a

variety of ways to enhance the flexibility or controllability

of power systems. One such application is the use of FACTS

devices to provide direct control over the amount of power

flowing in a particular transmission line, or group of lines,

in an interconnected AC system [2,3]; this application has

variously been described as power scheduling [2,4], power

flow control [5,6], or closed-loop control of AC power flow

[7]. Such closed-loop control of power flow in an AC systemcan provide a number of possible benefits: preventing

unwanted loop flows in an interconnected system; allowing

power to be directed along a contract path in a transmission

system; preventing inadvertent overloading of lines already

near their thermal limits [2,6].

A FACTS devices capability to direct the flow of power

rests on its ability to control dynamically one or more

of the factors that influence power transfer in the line it

compensates. Thus, closed-loop power flow control can

be achieved using a thyristor controlled series capacitor

(TCSC), a static synchronous series compensator (SSSC),

or a unified power flow controller (UPFC), and a number

of schemes employing these different devices have

been proposed [3,4,5]. However, few researchers have

considered the possible impact of closed-loop power flow

control strategies on the stability characteristics of the rest

of the power system, or the influence of the response time

of these controllers on system stability, despite a range of

powerflow controller designs and controller response times

having been proposed in the literature.

References [2] and [4] suggest that the response time

of a closed-loop power flow controller should be on the

order of 10 to 30 seconds, whereas references [3] and [5]

report on controller designs with response times of tens

of milliseconds. Reference [4] does consider the impact

of FACTS device control on the stability characteristics of

the power system as a whole, but in that study the FACTS

device was equipped with both a power flow controller and

a stabilising damping controller acting simultaneously:

as such, it is not possible to draw conclusions from [4]

on the impact of their power flow control strategies, or

the response times of the power flow controller itself, in

isolation from the other (supplementary) control functions

of their FACTS device.

This paper examines the impact of closed-loop power

flow control on the small-signal and large-signal stabilitycharacteristics of an AC power system in isolation from

any other supplementary controllers such as FACTS power

oscillation damping controllers or power system stabilisers

this is not to imply that such supplementary controls

would not or should not be present in a power system,

merely that the objective of the paper is to focus on, and

isolate the influence of closed-loop power flow controllers

on system stability. The paper considers a single generator

infinite bus study system in order to allow the fundamental

interactions to be examined readily, and considers a TCSC

as the FACTS controllable compensator at the heart of the

power flow controller. Furthermore, the paper considers

power flow controller designs whose response times are

on the order of several seconds as considered in [2, 4].

Two distinct strategies have been proposed for implementing

closed-loop powerflow control. The first, constant power

strategy [2,4] involves forcing the uncompensated line (or

lines) in a transmission system to absorb any increase in

the power dispatched, while the second, constant angle

strategy [2,4] regulates the flow of power along a particular

line in which the controllable series compensation is

applied. This paper presents a theoretical analysis of both

the constant power and constant angle control strategies

Copyright (c) 2004 IEEE. This paper was first published in AFRICON 04,

15-17 September 2004, Gabarone, Botswana


2/9

Vol.97(1) March 2006 SOUTH AFRICAN INSTITUTE OF ELECTRICAL ENGINEERS 35

as well as the basic philosophy behind these two control

strategies in an interconnected AC transmission system.

The paper examines not only the impact of a power flow

controllers response time on the small-signal and large-

signal stability characteristics of a power system, but in

addition examines whether the mode of power flow control

(i.e. constant power vs. constant angle) itself has anyimpact on stability characteristics.

Figure 1: Single-line diagram of the study system.

2. DETAILS OF STUDY SYSTEM

2.1 System Overview

Fig.1 shows a diagram of the single-generator study-

system considered in this paper. The system consists of

a synchronous generator that is connected to an infinite

busbar via a transformer and two parallel transmission

lines. Transmission line L1 is compensated with a TCSC,

while line L2 is uncompensated.

The structure of the study system shown in Fig.1 is based on

that which was used to study line power scheduling in [4].However in this paper the parameters of the study system

are different from those in [4], and are based on those of

the Machines Research Laboratory at the University of

KwaZulu-Natal [9]. In addition, in this study the generator

at the sending end of the transmission line is equipped

with an automatic voltage regulator (AVR). The inclusion

of the generators AVR is important when considering the

stability characteristics of a study system, since an AVR is

known to have a significant impact on both small-signal

and transient stability characteristics [11]. A detailed

simulation model of the study system shown in Fig.1 has

been developed in the power system simulation package

PSCAD [13]. The key elements of this study system model

are discussed briefly below.

The synchronous generator is represented using a detailed

(7th-order) electro-mechanical model within PSCAD. The

generator is connected to an infinite busbar via two parallel

transmission circuits and a transformer. Each transmission line

is represented using lumped impedances, while the generator

step-up transformer is represented by its leakage reactance.

2.2 TCSC Model

Fig.2 shows a single-line diagram of a TCSC, which

comprises a capacitor in parallel with a thyristor-controlled

reactor (TCR). This device is inserted in series with the

transmission line, much like a series compensating

capacitor. The net compensating reactance -jXTCSCthat theTCSC provides to the system is the parallel combination of

itsfixed capacitive reactance -jXCand the variable inductive

reactance, jXTCR

of its TCR, where the latters magnitude is

a function of the thyristor delay angle . For the purpose of

interfacing a TCSC to a high-level controller, the devices

control input is not the thyristor delay angle but rather the

TCSCs reactance order Xorder

, where

(1)

Figure 2: Single-line diagram of the TCSC and its internal

controls.

The reactance order Xorder

of a TCSC is thus a dimensionless

ratio (gain) that defines the extent to which the devices

net compensating reactance is increased over the value of

its fixed internal capacitive reactance XC; for a practical

TCSC, the value of Xorder

can range between 1 and 4 [10].

In the PSCAD model of the TCSC used in this study, a

linearization function, in the form of a look-up table for

Xorder

to thyristor firing angle () mappings, is used to

calculate the correct value of for the demanded Xorder

value at the input to the TCSC [7].

The PSCAD model of the TCSC used in this study

represents the individual components of the device in all

three phases, including its power electronic switches and

their low-level firing controls. Although thefiring angleof

a TCSCs thyristors is measured from the zero crossings of

its capacitor voltage VC, in practice the synchronization of

the thyristor firing controls is usually carried out indirectly

by means of a phase locked loop (PLL) synchronized to


3/9


the transmission line currents [12] in order to ensure stable

operation of the TCSC. The PSCAD model of the TCSC

used in this study includes a detailed representation of a

phase locked loop which calculates the instantaneous

angle of the TCSC voltages from the measured line

currents; the model also represents the low-level controls

used to generate thyristor firing signals in each phase bycomparing to . Finally, the PSCAD model includes a

surge arrester connected across each phase of the TCSC as

shown in Fig.2. Such surge arresters are always a feature

of TCSC installations in the field [10,12]; it has been found

necessary to include them in the TCSC model in transient

stability studies to ensure satisfactory performance of

the TCSCs capacitor voltages, and hence of its PLL-

based firing controls, following short-circuit faults in the

transmission line [14].

This PSCAD simulation model of the single generator

study system, together with the detailed TCSC model, was

used to investigate the performance of a closed-loop powerflow controller implemented in transmission line L1. The

theory of power flow control is discussed in the following

section.

3. POWER FLOW CONTROL THEORY

In Fig.1, the approximate expression for the active power

transfer in line L1 is given by

(2)

where |V2| and |V

3| are the magnitudes of the voltages at

buses 2 and 3 of the system and 23

is the transmission

angle between these bus voltages.

Equation 2 highlights the principle by which series

compensation can be used to manipulate the net impedance

of a particular transmission line and hence influence its

power transfer. When the TCSCs reactance is increased for

a given transmission angle 23

, the net reactance XLX

TCSC

of the compensated line is reduced, thereby increasing the

active power transfer. Hence, the magnitude of the active

power transfer in the compensated line can be increased

for a given transmission angle by increasing the amount

of TCSC compensation. Alternatively an increase in theamount of TCSC compensation can be used to reduce

the transmission angle required for a given active power

transfer in the compensated line.

These two observations underlie two distinct strategies

that have been proposed for closed-loop control of line

power flow [2]: the constant power strategy which

keeps the power flow in the compensated line constant,

and the constant angle strategy which ensures that the

compensated line transfers any increase in dispatched

power. Each of these strategies is now discussed in more

detail.

3.1 Constant Power Strategy

Consider the system shown in Fig.1 initially operating at

steady state, and then subjected to an increase in the output

power of the generator. This increase in the generator

output power (Pt) causes the common transmission angle

23

across both lines L1 and L2 to increase, and hence the

active power transfer across both lines initially increases.

However in the constant power strategy the TCSC is to be

used to keep the power flow in line L1 constant at some

desired set point value. The relationship in (2) shows that

to achieve this, the TCSCs capacitive reactance has to be

altered to counter any change in angle across line L1. In

other words, when the generator output power is increased,

the TCSCs reactance is then decreased accordingly so

that the power transfer in line L1 remains unchanged fromthe desired set point value. Consequently, in this mode

of control, all the increase in dispatched power from the

generator is then forced to flow through line L2.

3.2 Constant Angle Strategy

Consider once again, the situation when the generator

output power Ptin the system of Fig.1 is increased, but now

the TCSC is to be used to maintain a constant transmission

angle23

across both lines. From (2) in order for additional

power to be transmitted, either the capacitive reactance of

the TCSC has to be increased or the transmission angle 23

has to be increased. In this scenario, where the angle acrosslines L1 and L2 is to be kept constant, when the generator

output power increases the TCSCs capacitive reactance

has to be increased accordingly. Increasing the TCSCs

reactance in this manner results in the compensated line

L1 transferring all the additional power dispatched. This

ensures that the angle across line L1 and L2 is kept constant

and that the power transfer in the uncompensated line L2

remains unchanged.

3.3 Structure of the Power Flow Controller

A feedback control system has been developed in order

to implement TCSC-based, closed-loop control of

transmission line power flow in the study system of Fig.1; the structure of this control system has been devised in

such a way that the powerflow controller can be operated in

either the constant power mode or the constant angle mode

simply by setting a toggle switch in the PSCAD simulation

model. Figs. 3 and 4 show the structure of this feedback

control system as it appears for each of the two settings

of the toggle switch. Note that the initial setting for the

powerflow controller is always constant power mode: that


4/9


is, the power flow controller starts, by default, in constant

power mode, and can be switched into constant angle mode

thereafter if desired.

Note also that in both power flow control modes, the

innermost part of the controller is the same in each case.

Specifically, in both Fig. 3 (constant power mode) and Fig.4 (constant angle mode) a feedback loop compares a signal

PL1

*, representing the commandedvalue of power transfer

in line L1, to the actual (measured) value of power transfer

in line L1 (PL1

) in order to generate an error signal PL1

;

this error signal PL1

is used to drive a proportional-integral

(PI) controller which adjusts the Xorder

value of the TCSC

around some set-point value Xorder 0

in order to force PL1

to

follow the commanded value PL1

*.

Figure 3: Block diagram showing the structure of the

powerflow controller as it appears in the constant power

mode of control.

Figure 4: Block diagram showing the structure of the power

flow controller as it appears in the constant angle mode ofcontrol.

The difference between the two mode settings of the power

flow controller lies in how this commanded value of PL1

*at

the input to the feedback loop is created in each case: the

actual value of PL1

*is obtained by adding the output of the

Mode Select switch to the value of a user-settable input

PL1 set

which represents the set-point value for the power

transfer in line L1.

Thus, in Fig. 3, when the Mode Select switch is set to

position A, the output of this Mode Select switch is zero

and the commanded value of power transfer in line L1

is then simply the set-point value PL1 set; in this case thefeedback loop will then maintain the power transfer P

L1in

line L1 at whatever value is chosen by the user at the input

PL1 set

, that is the controller will operate in constant power

mode.

Once the power flow controller has reached steady state in

constant power mode it can then, if desired, be switched

into constant angle mode by activating the toggle switch in

Fig. 4. When the toggle switch is activated in this way, itperforms two functions: firstly, it changes the Mode Selectswitch to position B; secondly, it activates a sample andhold circuit which takes a measurement of the active poweroutput P

tat the generator terminals and saves its value at

the onset of constant angle control to a storage variable Pt 0

.

Fig. 4 shows that a signal

Pt= Pt

Pt 0is then created whichrepresents the change in power dispatch from the generatorsince the time at which constant angle control commenced;in constant angle mode, (Mode Select switch in positionB) this signal P

tis then added to the set-point value P

L1 set

to form the new commanded value PL1

*of power transferin line L1 applied at the input to the feedback loop. In thisway, if and when the power flow controller is switched toconstant angle mode, the commanded input to the powerflow controller becomes P

L1

*= PL1 set

+ Pt, such that line

L1 is forced to transfer its initial power, plus any changeingenerator dispatch.

4. SMALL SIGNAL BEHAVIOUR

4.1 Verification of control modes

This section examines the impact of the two closed-looppowerflow control strategies on the small-signal dynamiccharacteristics of the study system in Fig.1. However, priorto considering this issue, the basic characteristics of thetwo modes of power flow control are verified by means ofthree simulation studies.

To test the performance of each control strategy, in eachcase the simulation study was started from the same steadystate condition in which the generators active power outputP

t= 0.769 p.u., and with this active power being initially

transferred by the two lines L1 and L2 as follows: PL1

=

0.462 p.u. and PL2= 0.307 p.u.

Figure 5: Response of the study system with no power

flow control.

Subsequently, the mechanical input power to the generatorwas increased by P

m= 0.04 p.u. such that the total active

power Pt dispatched was increased to 0.809 p.u. The

response of the system to this small increase in generatordispatch was then studied for the system without the power


5/9


flow control activated (i.e. with the TCSCs Xorder

held

constant) as well as with the power flow control activated

in each of the two control modes.

Fig. 5 shows the simulated response of the study system

to the increase in generator dispatch without power flow

control; the variables shown are the active power transfersin each line as well as the TCSCs X

ordervalue. The result

in Fig.5 confirms that, without the power flow controller

activated, the additional power dispatched by the generator

is shared between both of the two parallel transmission

lines, with line L1 transferring a larger proportion of the

additional power: line L1 transfers a larger share of the

dispatched power since it is electrically shorter than line

L2 as a result of the (in this case) fixed value of TCSC

compensation. In this scenario (fixed TCSC compensation)

any increase in generator dispatch results in both

transmission lines transferring a portion of the increased

power, and no form of power flow control along a contract

path can be instituted.

Fig. 6 now considers the study system with the power

flow controller active and set to the constant power mode.

The results show that, following the increase in generator

dispatch, initially the power transfer in both lines L1 and L2

increases, which then results in the power flow controller

reducing the degree of capacitive compensation provided

by the TCSC. By reducing the capacitive reactance of the

TCSC, the power transfer in line L1 decreases accordingly

and returns to its nominal operating point of 0.462 p.u. All

of the additional power dispatched by the generator is thus

forced to flow through line L2, where the power transfer is

increased to 0.347 p.u. This response thus satisfies the basic

philosophy of the constant power strategy discussed inSection 3, and confirms that the constant power controller

is correctly implemented in the PSCAD model of the study

system in Fig.1.

Figure 6: Response of the study system with the power

flow controller in constant power mode.

Fig.7 shows the behaviour of the study system following

the increase in generator dispatch with the power flow

controller now set to operate in constant angle mode.

According to [2,4] the constant angle strategy should

ensure that the compensated line transfers all the additional

output power of the generator. Fig.7 shows that initially,

following the increase in generator dispatch, the active

power transferred by both lines increases; however, the

power flow controller responds by increasing the degree

of capacitive compensation provided by the TCSC such

that the power transfer in line L2 returns to its nominal

operating point of 0.307 p.u and all the additional dispatch

is transferred by line L1 (PL1

increases to 0.502 p.u). Fig.7

thus confirms that the constant angle controller is correctly

implemented in the PSCAD model of the study system in

Fig.1.

4.2 Influence of power flow control on small-signal

characteristics of the study system.

Section 4.1 has confirmed the correct operation of the power

flow controller in the constant power and constant angle

modes. This section now examines the impact that each

of these power flow control modes has on the small-signal

stability characteristics of the study system when the power

flow controller is designed to respond at different rates.

Previous work [7] has described a technique for designing

the dynamic response characteristics of the closed-loop

powerflow controller. In this section, the design method of

[7] has been used to arrive at three different settling times

for the power flow controller implemented on the study

system of Fig.1. The settling times considered are, tS= 5s;

tS

= 10s and tS

= 25s. For each of these settling times, and

for both modes of power flow control, the response of the

study system was examined for a small step increase in the

mechanical input power to the generator.

Fig.8 compares the small-signal response of the study

system to an increase in generator dispatch for the three

different settling time designs of the controller in constant

power mode. In considering the responses in Fig. 8, there

are two aspects to the behaviour of the system that are

Figure 7: Response of the study system with the power flow

controller in constant angle mode.


6/9


affected by the design of the power flow controller. Firstly,

the rate at which the power transfer in each line is returned

to the correct post-disturbance steady state value is

different in each case, which is the expected (and intended)

consequence of adjusting the controller design. However

the results show that the damping of the generators

electromechanical oscillations is also affected by the powerflow controllers response time: close inspection of Fig. 8

shows that the rate of decay of the oscillatory components

of the line powers PL1

and PL2

becomes greater as the

settling time of the power flow controller is made shorter.

The electromechanical swing mode of the generator in

this system has a frequency of approximately 1 Hz. As

such, one would expect that the power flow controller

would be more likely to influence the characteristics of

this electromechanical swing mode (via the associated

oscillations in the line power transfers) as the response

of the controller is made faster. Conversely, one would

expect the power flow controller to have less influence

on the characteristics of the systems electromechanical

oscillations when its response time is designed to be

significantly longer than the period of these oscillations.

However, the results in Fig. 8 show not only that the

power flow controller has a greater influence on the

electromechanical oscillations of the study system as its

response time is made shorter, but in addition that this

influence is to increasethe damping of these oscillations,

at least in constant power mode.

Fig.9 once again compares the response of the system to

an increase in dispatch at the three different settling times

of the power flow controller, but with the controller now

operating in constant angle mode. As in the case of constantpower mode, the design of the controllers settling time

influences not only the rate at which the power transfers in

each line are returned to their correct steady state values,

but also the damping of the post-disturbance oscillatory

components present in these power transfers as a result of

the generators electromechanical swings.

Figure 9: Small-signal response for different powerflow controller

settling times: constant angle mode.

Close inspection of Fig. 9 shows that in constant angle

mode, the influence of the power flow controller is to

decrease the damping of the systems electromechanical

oscillations: the results show that as the settling time of

the power flow controller is made shorter, the systems

electromechanical oscillations take progressively longer to

die out following the disturbance.

The results of the small-signal investigations in this section

have thus shown that in both modes of operation, the

closed-loop power flow controller has a more pronounced

influence on the damping of the systems electromechanical

oscillations as its response time is made shorter. However

the natureof this influence of the power flow controller

is dependent on its mode of operation: in constant power

mode the influence is beneficial, adding to the inherent

damping of the system; in constant angle mode the influence

is detrimental, acting to diminish the systems inherent

damping. The reasons for the opposite influences on system

damping in the two power flow control modes requiresfurther investigation, but the conclusion is clear: operation

in constant angle mode is detrimental to the inherent

damping of the system, particularly at short settling times

of the power flow controller; as a result, careful design of a

constant-angle controller would be required, in conjunction

with any other damping controllers present in the system,

prior to practical implementation of such a scheme.

5. LARGE SIGNAL BEHAVIOUR

The previous section has considered the impact of power

flow controller design, and controller mode, on the small

signal characteristics of the study system. This sectionnow examines the impact of closed-loop power flow

control on the behaviour of the study system under the

transient conditions that typically follow a large system

disturbance.

Once again the characteristics of the study system in Fig.1

are investigated, firstly with the power flow controller

disabled, and thereafter for each of the two modes of power

flow control, in this case for a single value of settling time

tS= 5s. The study system in Fig.1 was started from a steady

state condition at which the generators total active power

output is 0.733 p.u. A three-phase short-circuit fault was

then applied in the uncompensated line L2 and removedafter 700ms, with the fault located 33% of the way along

the line from bus 2. The fault duration of 700 ms chosen

for this study is relatively long: although faults of this

duration can occur in practice, they are relatively rare.

However, this severe disturbance to the study system has

been chosen in order to be able to demonstrate more clearly

the influence of the different power flow controller modes

under transient conditions.


7/9


Fig. 10 shows the simulated response of the study system

following the three-phase short-circuit fault with no power

flow control. The variables shown are the generators

transmission angle relative to the infinite bus, its rotor

speed deviation, as well as the value of Xorder

at the input to

the TCSC. When such a fault occurs, the electrical output

power of the generator is reduced (since the fault is applied

midway along the line, there is still some power transfer

on the unfaulted line) while the mechanical input power

remains constant. As a result of the imbalance between the

mechanical input power and electrical output power of the

generator, the generator rotor starts to accelerate and its

transmission angle increases. When the fault is removed

the electrical output power increases abruptly, such that

the electrical output power of the generator then exceeds

the mechanical input power, hence causing the generator

to decelerate and, in this case, return to synchronism. Notealso, that with no power flow control implemented, the

amount of compensation provided by the TCSC remains

constant both during the fault, and in the post-fault transient

period.

Figure 10: Response of study system to the 3-phase short-circuit

fault: no power flow control.

Fig. 11 now shows the response of the study system with the

powerflow controller active in the constant power mode.

As before, when the fault is applied, the total output power

of the generator decreases causing the generator rotor to

accelerate. However, in this case, with the power flow

controller activated, the controller responds to the transient

swings in the power transferred by the lines by varying

the amount of compensation provided by the TCSC. In

particular, Fig. 11 shows that the response of the controllerin constant power mode is initially to increasethe TCSCs

Xorder

value in an effort to return the power transfer in the

compensated line L1 to its nominal operating point when

the fault appears. However once the fault is removed,

the electrical output power of the generator increases,

which results in the compensated lines power transfer

temporarily exceeding its nominal operating point value.

The power flow controller then decreases the TCSCs Xorder

accordingly in order to return the power in the compensated

line to its pre-disturbance operating point value.

Fig. 12 now shows the response of the study system with

the power flow controller in constant angle mode. Theresults show that in this mode of operation, the power flow

controller responds to the short circuit fault by initially

sharply decreasingthe compensation in line L1.

Recall that in constant angle mode the controller is designed

to respond to a reduction in generator dispatch by reducing

the degree of compensation of line L1 in order to maintain a

Figure 8: Small-signal response for different powerflow controller

settling times: constant power mode.

Figure 11: Response of the study system to the 3-phase short-

circuit fault: constant power mode.

Figure 12: Response of the study system to the 3-phase short-

circuit fault: constant angle mode.


8/9


constant angle across both lines. In this case, the reduction

in generator output power is a transient one caused by a

short-circuit fault, rather than as a result of a permanent

change in system operating point, but the constant angle

control nevertheless responds to the transient swings in the

generator and transmission line power transfers.

Comparison of Figs. 11 and 12 thus shows that the power

flow controller initially responds in the opposite manner

to the same short-circuit fault in its two different modes of

operation: in constant power mode the controller initially

responds by reducing the net impedance of the compensated

line, with the effect being that the total power transfer out

of the generator during the fault is increased by the action

of the controller; in constant angle mode, the controller

initially responds by increasing the net impedance of the

compensated line, with the effect being that the total power

transfer out of the generator during the fault is reduced by

the action of the controller.

It is well known that the transient (first-swing) stability

of a generator is enhanced by any action that results in

improved transfer of active power out of the generator

during, and immediately after a fault condition [11]. The

results in Figs. 11 and 12 therefore suggest that the mode

of the power flow controller will have an effect on the

first swing characteristics of the generator in response to a

short-circuit fault. Fig. 13 therefore compares the response

of the generators transmission angle and speed deviation

to the short-circuit fault for the cases where the power flow

control is inactive; active in constant power mode; active

in constant angle mode (i.e. the results of Figs. 10, 11 and

12 are now plotted on the same axes).

Figure 13: Comparison of responses to the short-circuit fault:

constant power; constant angle; no control.

The results in Fig. 13 confirm that the powerflow controller,

and its mode of operation, both influence the first swing

characteristics of the generator: the amplitude of the first

swing of the generator angle is slightly smaller in constant

power mode than for the case when there is no power flow

control; by contrast, the amplitude of the first swing of

the generator angle is noticeably larger when the power

flow controller is in constant angle mode. The post-fault

behaviour of the generators speed deviations in Fig.13

also provide further confirmation of the findings in Section4 of the paper the small-signal damping is significantly

improved by the power flow controller in constant power

mode whereas the damping is clearly reduced by the

controller in constant angle mode.

6. CONCLUSION

This paper has considered the application of a thyristor

controlled series capacitor for closed-loop control of

power flow in both constant power and constant angle

modes of operation. The results indicate that the power

flow controllers operation has an important influence on

both the small-signal and transient stability characteristicsof a power system. In particular it has been shown that

the constant angle mode of operation can be detrimental

to system damping and first swing stability, particularly

for a relatively fast-responding power flow controller. By

contrast, the constant power mode of operation has been

shown to have a beneficial impact on both system damping

and first-swing stability under the conditions studied.

7. REFERENCES

[1] Adapa R, Baker MH, Habashi K: Proposed Terms

and Definitions for Flexible AC Transmission System

(FACTS), IEEE Transactions on Power Delivery,

Vol.12, No.4, 1997.

[2] Larsen EV, Bowler CEJ, Damsky B, Nilsson S: Benefits

of Thyristor Controlled Series Compensation, CIGRE

Paper 14/37/38-04, Paris, 1992.

[3] Gyugyi L, Rietman TR, Edris A, The Unified

Power Flow Controller: A New Approach to Power

Transmission Control IEEE Transactions on Power

Delivery, Vol. 10, No.2, April 1995.

[4] Martins N, Paserba JJ, Pinto JCP: Using a TCSC

for Line Power Scheduling and System OscillationDamping Small Signal and Transient Stability

Studies, Proceedings IEEE PES Winter Meeting,

Singapore, January 2000.

[5] Mihalic R, Papic I, Power Flow Control Using Static

Synchronous Series Compensator, Proceedings UPEC

97, pp. 174-177.


9/9


[6] Noroozian M, Andersson G, Power Flow Control

By Use Of Controllable Series Components, IEEE

Transactions on Power Delivery, Vol. 8, No.3, July

1993, pp. 1420-1429.

[7] Rigby BS: An AC Transmission Line Power Flow

Controller Using a Thyristor Controlled SeriesCapacitor, Proceedings IEEE Africon 2002, George,

South Africa, October 2002, pp.773-778.

[8] Ally A, Rigby BS, The Application of a Thyristor

Controlled Series Capacitor for Closed-Loop Control

of Transmission Line Power Flow, Proceedings

SAUPEC 04, Stellenbosch, South Africa, Jan 2004.

[9] Limebeer DJN, Harley RG, Schuck SM,

Subsynchronous Resonance of the Koeberg turbo-

generators and of a laboratory micro-alternator

system, Transactions of the SAIEE, Vol. 70, Part 11,

Nov. 1979, pp. 278-297.

[10] Hingorani NG, Gyugyi L, Understanding FACTS:

Concepts and Technology of Flexible AC Transmission

System, IEEE Press, Piscataway New Jersey, 1999,

ISBN 0-7803-3455-8.

[11] Kundur P, Power System Stability and Control,

McGraw Hill Inc, New York, 1994.

[12] Song YH, Johns AT, Flexible AC Transmission

Systems, IEEE Power & Energy Series 30, ISBN 085

2 9677 13.

[13] Manitoba HVDC Research Centre Inc: Introduction

to PSCAD/EMTDC Version 3.0.

[14] Carpanen RP, Rigby BS: Transient Stability

Enhancement Using a Thyristor Controlled Series

Capacitor, Proceedings IEEE Africon 2004,

Gaborone, Botswana, September 2004.

8. APPENDIX

Line Parameters (L1& L

2): R

L= 0.033 pu X

L= 0.75 pu

TCSC Parameters: XC= 0.124 pu XL = 0.025 puTransformer Parameters: X = 0.13 pu

Synchronous Generator Parameters documented in reference [9]

Per Unit System: Vbase

= 220 Vrms (l-l) Pbase

=3KVA

Zbase

= 16.13 Ohm Ibase

=7.87Arms

Documents

The Impact of Closed-loop Power Flow Control Strategies on Power System Stability Characteristics in a Single-generator System-2006