Excitation and Governing Control of a Powern Generation Based Intelligent System

Abstract
Modern power systems are complex and non-linear and their operating conditions
can vary over a wide range. In this work, the power system (PS) transient terminal
voltage and frequency stability enhancement have been well investigated and studied
through the following efforts.
• Enhancing the responses of the transient stability by adopting conventional PID
controllers as an additional voltage controller with the Automatic Voltage Regulator
(AVR) in the excitation system for terminal voltage, and in the governing system for
frequency deviation response.
• ANN (NARMA-L2) system is proposed as an effective controller model to achieve the
desired enhancement. This model after training can be called as (Identifier). This
identifier follows the system behavior even in situation of high disturbances.
There are enhancement progress in terminal voltage Vt , and frequency deviation Δω
through the investigation for the three cases (without controller, with PID controller, and
with NN controller) for single machine infinite bus using MATLAB – Simulink software.
Keywords: PID controller, Neural Network controller, Excitation system control,
Governing system control.
السيطرة على منظومة الاثارة و التحكم في وحدة توليد القدرة بواسطة
منظومة ذكية
الخلاصة
تعتبر أنظمة القدرة الحديثة لاخطية وعالية التعقيد وان شروط اشتغالها يمكن أن يتغير ضمن
مجال واسع. يقدم هذا الجهد تقصي ودراسة استقرارية نظام القدرة وذلك بتحسين الأستقرارية العابرة
للفولتيات والتردد بشكل تام ومن خلال الخطوات التالية.
• تحسين هذه الأستقرارية من خلال تبني المسيطر(التناسبي – التكاملي – التفا ضلي) التقليدي وذلك
في منظومة الإثارة لتحسين فولتية ( AVR ) كمسيطر احتياطي يعمل مع منظم الفولتية الأوتوماتيكي
الأطراف وفي منظومة التوربين لتحسين انحراف التردد.
والذي (NARMA-L • لقد تم اقتراح الشبكات العصبية الاصطناعية كنموذج مسيطر من نوع ( 2
يمكن ان يدرب بشكل دقيق لكي يكون إخراج هذا المسيطر هو الإخراج الحقيقي للمنظومة (المطابق)
ويتم تحسين الأداء بتتبع سلوك المنظومة حتى وان كان الأضطراب كبير .
لقد تم تحقیق تحسین استجابة فولتیة الأطراف والانحراف في التردد عن طریق مقارنة وتطبیق ثلاث
حالات ( بدون استخدام أي مسیطر ,باستخدام المسیطر(التناسبي – التكاملي – التفا ضلي) , و أخیرا
للمحاكاة وطبق (MATLAB) باستخدام مسیطر الشبكة العصبیة الاصطناعیة حیث تم اعتماد برنامج
على مولدة أحادیة مختلفة الأحمال في منظومة الشبكة الكھربائیة.
PDF created with pdfFactory Pro trial version pdffactory.com
Eng. & Tech. Journal, Vol. 28, No. 5, 2010 Excitation and Governing Control of a Power
Generation Based Intelligent System
1- Introduction
Power system stability can be defined as
the tendency of power system to react to
disturbances by developing restoring
forces equal to or greater than the
disturbing forces to maintain the state of
equilibrium (synchronism). Stability
problems are therefore concerned with
the behavior of the Synchronous
Generator (SG) after they have been
perturbed. Generally, there are three
main categories of stability analysis.
They are namely steady state stability,
transient stability and dynamic stability.
Steady state stability is defined as the
capability of the power system to
maintain synchronism after a gradual
change in power caused by small
disturbances [1]. Transient state stability
refers to as the capability of a power
system to maintain synchronism when
subjected to a severe and sudden
disturbance [2]. The third category of
stability, which is the dynamic stability,
is an extension of steady state stability, it
is concerned with the small disturbances
lasting long period of time. The
generators are usually connected to an
infinite bus where the terminal voltages
(Vt) are held at a constant value.
The study of SG control systems can
roughly be divided into two main parts:
voltage regulation and speed governing.
Both of these control elements contribute
to the stability of the machine in the
presence of perturbations. There are
various methods of controlling a SG and
suitability will depend on the type of
machine, its application and the operating
conditions[3].
The AVR system has a substantial effect
on transient stability when varying the
field voltage to try to maintain the
terminal voltage constant. This is
achieved by comparing the output voltage
With a reference voltage and, from the
difference, it makes the necessary
adjustments in the field current to bring
the output voltage closer to the required
value [3]. The excitation and governing
controls of the generator play an
important role in improving the dynamic
and transient stability of the power
system.
The presence of poorly damped modes
of oscillation, and continuous variation in
power system operating conditions arises
some limitations in the conventional
controllers. These limitations have
motivated research into so-called
intelligent control systems. Artificial
Neural Networks (ANNs) have been used
in the design of nonlinear adaptive
controllers with various control objectives
in the field of electrical power
engineering, especially for the
synchronous generator excitation and
governor control. The NARMA-L2
controller among the ANN family has in
particular attracted attention for the
control of SG and flexible ac transmission
systems devices due to its powerful
control capability for damping of lowfrequency
oscillations as well as the faster
convergence speed for identifying the
plant[4]. This research is focused mainly
on voltage and frequency stability of SG
in a typical power system using fourth
order model of synchronous generator.
2- Modeling of Synchronous Generator
The overall accuracy of the
power system stability is primarily
decided by how correctly the Synchronous
Generators within the system are
modeled. The proposed simulation model
PDF created with pdfFactory Pro trial version pdffactory.com
Eng. & Tech. Journal, Vol. 28, No. 5, 2010 Excitation and Governing Control of a Power
Generation Based Intelligent System
is developed as a fourth order machine
time constants in order to improve the
terminal voltage and frequency deviation
responses [5]. With proper modeling of
the synchronous machine in the power
system, a better understanding of how the
machine reacts under sudden large
disturbances during transient conditions
can be achieved and hence a better power
system voltage regulator and governor
controllers of the SG can be designed .
Some assumptions were taken into
consideration and made prior to the design
of the simulation model, these
assumptions are:
-The SG turbine in this model produce a
constant torque with a constant speed
maintained during steady state operation.
-The SG output terminals are connected to
infinite bus bar that has various load
changes.
- Only basic and linear models of the
power system components will be used.
-All the time constants of the SG which
are used in this model of all components
are assumed to be the optimum time
constants extracted based on the values
given in Walton[5].
The stability of a SG depends on the
inertia constant and the angular
momentum. The rotational inertia
equations describe the effect of unbalance
between electromagnetic torque and
mechanical torque of individual machines.
By having small perturbation and small
deviation in speed, the swing equation
becomes [1]:
dDω / dt = (1/2H) (DPm - DPe) ….. (1)
then dDω / dt = d²δ / dt
H = inertia constant
DPm = change in mechanical power
DPe = change in electrical power
Dω = change in speed (elec. rad/sec)
δ = rotor angle (rad.)
Using Laplace Transformation, equation
(1) becomes:
dδ / dt =Dω(s) = (1/2Hs) [DPm(s)- DPe(s)]
…… (2)
A more appropriate way to describe
the swing equation is to include a
damping factor that is not accounted for in
the calculation of electrical power Pe.
Therefore, a term proportional to speed
deviation should be included. The speedload
characteristic of a composite load
describing such issue is approximated by
[6]:
ΔPe=ΔPL+ KD Δω ….. (3)
where KD is the damping factor or
coefficient in per unit power divided by
per unit frequency. KDDω is the
frequency-sensitive load change and DPL
is the non frequency-sensitive load
change.
Figure (1) presents a block diagram
representation of a load change derived
from the swing equation with the aid of
equation (3) or:
Δω(s) = [ΔPm(s) - ΔPL(s)] [1/(2Hs +
KD)] ……(4)
Figure (2) represents a simplified block
diagram of the Governor and AVR of the
synchronous generator with the two
feedback quantities ( voltage and
frequency ).
The main control function of the
excitation system is to regulate the
generator terminal voltage Vt which is
accomplished by adjusting the field
voltage with respect to the variation of Vt
[7].
The following proposed models are
needed to study the effect of using the
PID controllers and the Neural Network
(NN) controller which represent on the
PDF created with pdfFactory Pro trial version pdffactory.com
Eng. & Tech. Journal, Vol. 28, No. 5, 2010 Excitation and Governing Control of a Power
Generation Based Intelligent System
fourth order model of synchronous
generator for terminal voltage and
frequency deviation stability control and
how this stability have been enhanced.
2.1 Generator Model
A fourth order model of the SG
consists of a generator gain plus four pairs
of pole-zero time constants can be
modeled. In terms of expressing this as a
transfer function, then the following
equation is given[3]:
…. (5)
There are two ways in MATLABSimulink
to design the machine model,
these are:
1. Using power system block set which is
a set of ready-made [4].
2. Using blocks of transfer functions of
the machine to manipulate the design
model.
However, using blocks of the transfer
function to represent the components in
the power system is capable of having
higher order machine time constants as
inputs. This can be achieved by the
illustration shown in Figure (3) [8,9].
Where: KG = Gain of the generator, Tz =
Time constant of the zero, Tp = Time
constant of the pole, VF = Field voltage of
the SG, Vt = Terminal voltage of the SG.
2.2 Exciter model
The most basic form of expressing the
exciter model can be represented by a gain
KE and a single time constant TE :
VF(s) / VR(s) = KE / (1 + sTE) ….. (6)
VR = the output voltage of the regulator
(AVR), VF = field voltage
The excitation system amplifier is
represented similarly by a gain KA and a
time constant TA. The transfer function of
the amplifier is:
VR(s)/ ΔVe(s)=KA / (1+ sTA) ……(7)
Where: ΔVe = Voltage error = reference
voltage (Vref ) - output voltage of the
sensor(VS).
2.3 Sensor Model
The terminal voltage of the SG is being
fed back by using a potential
transformer that is connected to the bridge
rectifiers. The sensor is also being
modeled, likewise as the exciter:
VS(s) / Vt(s) = KR / (1 + sTR) ...... (8)
VS = output voltage of the sensor, KR and
TR are the gain and time constant of the
sensor .
2.4 Automatic Voltage Regulator
(AVR) Mode
In most modern systems, the AVR is a
controller that senses the generator output
voltage then initiates corrective action by
changing the exciter control in the desired
direction [10].
A simple AVR is created with a 1st
order model of SG as shown in the Figure
(4).
From this block diagram, the closedloop
transfer function of a 1st order
relating the generator terminal voltage
Vt(s) to the reference voltage Vref(s) can be
written as follow:
A E G R A E G
A E G R
t ref sT sT sT sT KKK
Vs V s KKK sT
+ + + + +
+
=
(1 )(1 )(1 )(1 )
( ) / ( ) (1 )
1
……. (9)
2.5 Turbine Model
The simplest form of model for a nonreheat
steam turbine can be approximated
by using a single time constant TT. The
model for turbine associates the changes in
mechanical power DPm with the changes in
steam valve position D€V is given as:
(1 ) (1 ) (1 ) (1 )
( )/ ( ) (1 ) (1 ) (1 ) (1 )
1 2 3 4
1 2 3 4
P P P P
Z Z Z Z
t F G sT sT sT sT
Vs V s K sT sT sT sT
+ + + +
+ + + +
=
PDF created with pdfFactory Pro trial version pdffactory.com
Eng. & Tech. Journal, Vol. 28, No. 5, 2010 Excitation and Governing Control of a Power
Generation Based Intelligent System
GT(s) = DPm(s) / D€V(s) = 1 / 1 + sTT ..(10)
2.6 Governor Model
The speed governor mechanism works
as a comparator to determine the difference
between the reference set power DPref and
the power (1/R)Dw as shown in Figure (5).
The speed governor output DSg is therefore:
DSg(s) = DPref(s) – (1/ R )Dw(s) ….. (11)
where R represents the drop. Speed
governor output DSg is being converted to
steam valve position D€V through the
hydraulic amplifier. Assuming a
linearized model with a single time
constant Tg:
D€V(s) = (1 / (1 + sTg)) DSg(s) ….. (12)
The final simulation model for a 4th order
SG can be developed in "Matlab" as
shown in Figure (5).
Typically the excitation control and
governing control are designed
independently since there is a weak
coupling between them, then the voltage
and frequency controls are regulated
separately.
The suggested conventional PID
controller that can be used to enhance the
output response of the AVR in the
excitation system is differ from the
conventional PID controller that can be
used to enhance the frequency deviation
in the governing system.