Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
4Activity
0 of .
Results for:
No results containing your search query
P. 1
PID Tuning

PID Tuning

Ratings: (0)|Views: 1,168|Likes:

More info:

Published by: Satya Sai Babu Yeleti on Jan 29, 2011
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

02/22/2014

pdf

text

original

 
 
Tuning a PID Controller for a Digital Excitation Control System
Kiyong Kim and Richard C. Schaefer Basler ElectricRoute 143, Box 269Highland, IL 62040
Abstract — Some of the modern voltage regulatorsystems are utilizing the Proportional, Integral, andDerivative (PID) control for stabilization. Two PIDtuning approaches, pole placement and pole-zerocancellation, are commonly utilized for commissioning adigital excitation system. Each approach is discussedincluding its performance with three excitationparameter variations [1]. The parameters consideredinclude system loop gain, uncertain exciter timeconstants, and forcing limits. This paper is intended forvarious engineers, and technicians to provide a betterunderstanding of how the digital controller is tunedwith pros and cons for each method.
I. INTRODUCTION
Today’s digital excitation systems offer numerousbenefits for performance improvements and tuningover its analog voltage regulator predecessor. Thedays of potentiometers, screwdriver, and voltmetersfor tuning are replaced with a laptop computer . Unlikethe past, when excitation systems were tuned byanalog meters, today’s excitation can be tuned veryprecisely to desired performance and recorded into afile for future performance comparison in the form of an oscillography record.
II. PID CONTROL
The present day digital regulator utilizes a PIDcontroller in the forward path to adjust the response of the system. For main field excited systems, thederivative term is not utilized. The proportional actionproduces a control action proportional to the error signal. The proportional gain affects the rate of riseafter a change has been initiated into the control loop.The integral action produces an output that dependson the integral of the error. The integral response of acontinuous control system is one that continuouslychanges in the direction to reduce the error until theerror is restored to zero. The derivative actionproduces an output that depends on the rate of change of error. For rotating exciters, the derivativegain is used which measures the speed of the changein the measured parameter and causes anexponentially decaying output in the direction toreduce the error to zero. The derivative term isassociated with the voltage overshoot experiencedafter a voltage step change or a disturbance. Thebasic block diagram of a PID block utilized in theautomatic voltage regulator control loop is shown inFig. 1. In addition to PID block, the system loop gain(K
G
) provides an adjustable term to compensate for variations in system input voltage to the power converting bridge. When performance is measured,the voltage rise time is noted at the 10% and 90%level of the voltage change. The faster the rise time,the faster the voltage response [5].The benefits of a fast excitation controller can improvethe transient stability of the generator connected tothe system, or stated another way, maximize thesynchronizing torque to restore the rotor back to itssteady state position after a fault. A fast excitationsystem will also improve relay tripping coordinationdue to the excitation systems’ ability to restoreterminal voltage quickly, and providing more faultcurrent to protective relays for optimum tripping time.
K
P
K
I
s
Σ
V
REF
G
sK
D
1 + sT
D
Σ
++V
C
+-V
P
Σ
++
Π
V
RLMT
-V
RLMT
 / V
P
 / V
P
V
R
 V
REF
is generator voltage ReferenceV
C
is sensed voltageV
RLMT
is max field forcingV
p
is power input voltageV
R
is voltage regulator outputFig. 1: Simplified Block Diagrams of AutomaticVoltage Regulators
III. CHARACTERISTIC OFEXCITATION CONTROL SYSTEMS
An optimally-tuned excitation system offers benefits inoverall operating performance during transientconditions caused by system faults, disturbances, or motor starting [5]. During motor starting, a fastexcitation system will minimize the generator voltagedip and reduce the I2R heating losses of the motor.After a fault, a fast excitation system will improve thetransient stability by holding up the system andproviding positive damping to system oscillations.
 
 
A fast excitation system offers numerous advantages,improved relay coordination and first swing transientstability, however an excitation system tuned too fastcan potentially cause MW instability if the machine isconnected to a voltage weak transmission system.For these systems a power system stabilizer may berequired to supplement machine damping.The evaluation of system performance begins byperforming voltage step responses to examine thebehavior of the excitation system with the generator. Itis performed with the generator breaker open, sincethe open circuited generator represents the leaststable condition, i.e. the highest gain and the leastsaturation (See Fig. 2). Fig. 3 represents a bode plotof the generator and exciter field for a sweepfrequency from 0-10 Hertz. The frequency plotrepresents the potential voltage oscillation frequencyafter a disturbance. The time constant of thegenerator field and the exciter field is plotted andillustrated to show that as the frequency increases,the phase angle becomes more lagging. The phaseangle of the generator field adds directly with thephase angle of the exciter field. As the phase anglesadd to 150 degrees, the system will most likelybecome unstable because of the combined gain of the generator and the excitation system. Unlesscompensated properly through the PID controller, thesynchronous machine may become oscillatory after afault [8].Besides the open circuit voltage step test, another test performed is the voltage step test with thegenerator breaker closed. When voltage step testsare performed with the generator breaker closed, verysmall percentage voltage steps are introduced toavoid large changes in generator vars. In this case, a1-2% voltage step change is typical [7].Fig. 2: Generator Saturation Curve IllustratingGenerator GainFig. 3: Phase Shift of the Exciter Field, the Generator Field, and the Sum of the Two
IV. TUNING OF PID CONTROLLER
The controller parameters are determined withseveral excitation system parameters, such asvoltage loop gain and open circuit time constants [1].These parameters vary with not only the systemloading condition but also gains dependent on thesystem configuration, such as the input power voltagevia PPT to the bridge rectifier as shown in Figure 4.Commissioning a new AVR can be a challenging taskof checking excitation system data in a short time,without any test data, and with no other link to theactual equipment, except for an incompletemanufacturer’s data sheet, or some typical data set.Fig. 4: One-line DiagramTo tune the digital controller, two methods arepredominantly used, one being the pole placementmethod and the other being the pole zero cancellationapproach [1, 2]. To simplify the design of the PIDcontroller, we assume T
D
=0 in Fig. 1.Every PID controller contains one pole and two zeroterms with low-pass filter in the derivative blockignored. For generators containing rotating exciters,the machine contains two open loop poles, onederived from the main field and the other derived fromthe exciter field. A pole represents a phase lag in thesystem while the zero tends to provide a phase lead
 
 
component. The location of poles and zeros withrelation to the exciter and generator field polesdetermines the performance of the excitation controlsystem.Root locus is used to describe how the systemresponds based upon gain in the system. Using thepole placement method and referencing Fig. 5a, thepoles of the generator main field and exciter field arelocated on the real axis. The generator main field poleis located close to the origin while the exciter fieldpole is typically tens times the distance, the distancedepending upon the time constant of the exciter fieldversus the main field. The smaller the exciter fieldtime constant, the greater the distance. The PIDcontroller consists of one pole and two zeros.Fig. 5a shows how the closed loop poles move as theloop gain increases. The loop gain, K
G,
represents thetotalized gain that includes the generator, field forcingof the excitation system and PID controller. The openloop zeros becomes the closed loop zeros and do notdepend on the loop gain. On the other hand, the polesof the closed loop system (exciter, generator,controller) are moving toward zeros in a certain pathas the loop gain increases. The path of the closedloop poles depends on the relative location of polesand zeros based upon its time constants. With fixedPID gains, a certain system gain (K
G
) determines theclosed loop poles. Two cases of the closed poles areshown in Fig 5a, one for system gain 1.0 and theother for system gain 0.13.In general, the poles nearest to the origin determinethe system responses. When the poles become aconjugate pair, the system response will beoscillatory. The conjugate pair represents the ratio of the imaginary to real value of poles to determine thevoltage overshoot. Absolute value of pole determinesthe frequency of voltage oscillation. The faster voltageresponse can be achieved by moving the poles fromthe origin and the less oscillatory response with thesmaller ratio. See Fig. 5b.(a) Root Locus of Different Loop GainsGains: KP=90; KI=70; KD=25(b) Step Responses PerformanceGains: KP=90; KI=70; KD=25Fig. 5: Root Locus and Voltage Response with TwoDifferent System Gains (Kg)
Pole Placement:
In the pole placement method, thedesired closed-loop pole locations are decided on thebasis of meeting a transient response specification.The design forces the overall closed-loop system tobe a dominantly second-order system. Specifically weforce the two dominant closed-loop poles (generator and controller) to be complex conjugate pair resultingin an underdamped response. The third pole (exciter)is chosen to be a real pole and is placed so that itdoes not affect the natural mode of the voltageresponse. The effect of zeros to the transientresponse is reduced by a certain amount of trial anderror and engineering judgment involved in thedesign.The pole placement method generally requiresspecific information of the exciter field and maingenerator field time constants to determine the gainsneeded for the digital controller for adequateresponse. Voltage overshoot of at least 10-15% isanticipated with the pole placement method with a 2to 3 second total voltage recovery time, although itsvoltage rise time can be less then one second.Fig. 6 illustrates the generator terminal voltageperformance of a 100 MW steam turbine generator when a +5% open circuit voltage step change hasbeen introduced. Generator voltage overshoot is 20%with a total voltage recovery time of 2.5 seconds. ThePID gains are as follows: K
P
=90, K
I
=70, and K
D
=29.The excitation system bandwidth is used tocharacterize the response of the generator with thevoltage regulator. The wider the voltage regulator bandwidth, the faster is the excitation system.To derive the voltage regulator bandwidth, the gainand phase shift is plotted over a range of frequenciestypically 0-10 Hertz by applying a signal oscillator intothe voltage regulator summing point. The input signal

Activity (4)

You've already reviewed this. Edit your review.
1 thousand reads
1 hundred reads
Mudassar Hayat liked this
daljeetsinha2011 liked this

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->