Reactive power compensation in a steel industrial plant with several operating electric arc furnaces utilizing open-loop controlled TCR/FC compensators Arash Dehestani Kolagar * ,y , Arash Kiyoumarsi, Mohhamad Ataei and Rahmat Allah Hooshmand Department of Electrical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran SUMMARY Electric arc furnaces (EAFs) produce voltage fluctuations and flicker because of the reactive power severe variations. Furthermore, these loads absorb a large amount of reactive power. The static VAr compensators (SVCs) have been widely used by the industrial customers with arc furnaces to compensate the reactive power due to the quick response of the power electronic devices. In this paper, reactive power compensation in the steel industrial plant with several EAFs by utilizing open-loop controlled thyristor controlled reactor/fixed capacitor (TCR/FC) compensator is performed. The TCR/ FC compensator is usually applied in conventional steel making plants; one is in Mobarakeh/ Isfahan, Iran which is considered as the case study in this paper. Simulation results show that, although open-loop controlled TCR/FC is effective for compensating reactive power, it cannot efficiently compensate the fluctuations of the reactive power and reduce the flicker intensity. Copyright # 2010 John Wiley & Sons, Ltd. key words: electric arc furnace; open-loop controlled TCR/FC; flicker intensity 1. INTRODUCTION The electric arc furnace (EAF) is considered as a nonlinear and erratic load; it creates a series of adverse effects to the power grid such as three-phase unbalance of the power grid, negative sequence current, high order harmonics, severe voltage distortion, serious flicker, and low power factor. EAF compensation is usually performed using static VAr compensators (SVCs). Though, these devices are effective in mitigating flicker whenever controlled properly, their performance is limited due to inherent delays and their harmonics generation. A control method for the thyristor controlled reactor/fixed capacitor (TCR/FC) was presented to improve the dynamic response using a fast detection method [1]. This method is based on the substitution of conventional filtering by an efficient, faster, and simpler process with a lower computational burden and it guarantees that the compensation process will perform within 1 = 4 cycle of the fundamental period of the line voltage. On the other hand, a solution of obtaining an optimum flicker reduction at a comparatively low compensator power rating by an anti-windup extension of the controller is presented in Reference [2]. The dimensioning of a SVC for a steel plant grid with several operating EAF, taking the economic dimensioning of the compensator into account, is also demonstrated in this paper. EUROPEAN TRANSACTIONS ON ELECTRICAL POWER Euro. Trans. Electr. Power 2011; 21:824–838 Published online 21 July 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etep.479 *Correspondence to: Arash Dehestani Kolagar, Department of Electrical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran. y E-mail: [email protected]Copyright # 2010 John Wiley & Sons, Ltd.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
EUROPEAN TRANSACTIONS ON ELECTRICAL POWEREuro. Trans. Electr. Power 2011; 21:824–838Published online 21 July 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etep.479
Reactive power compensation in a steel industrial plant withseveral operating electric arc furnaces utilizing open-loop
controlled TCR/FC compensators
*C
UnyE-
Co
Arash Dehestani Kolagar*,y, Arash Kiyoumarsi,Mohhamad Ataei and Rahmat Allah Hooshmand
Department of Electrical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran
SUMMARY
Electric arc furnaces (EAFs) produce voltage fluctuations and flicker because of the reactive powersevere variations. Furthermore, these loads absorb a large amount of reactive power. The static VArcompensators (SVCs) have been widely used by the industrial customers with arc furnaces tocompensate the reactive power due to the quick response of the power electronic devices. In thispaper, reactive power compensation in the steel industrial plant with several EAFs by utilizing open-loopcontrolled thyristor controlled reactor/fixed capacitor (TCR/FC) compensator is performed. The TCR/FC compensator is usually applied in conventional steel making plants; one is in Mobarakeh/ Isfahan,Iran which is considered as the case study in this paper. Simulation results show that, although open-loopcontrolled TCR/FC is effective for compensating reactive power, it cannot efficiently compensatethe fluctuations of the reactive power and reduce the flicker intensity. Copyright # 2010 John Wiley &Sons, Ltd.
key words: electric arc furnace; open-loop controlled TCR/FC; flicker intensity
1. INTRODUCTION
The electric arc furnace (EAF) is considered as a nonlinear and erratic load; it creates a series of
adverse effects to the power grid such as three-phase unbalance of the power grid, negative sequence
current, high order harmonics, severe voltage distortion, serious flicker, and low power factor. EAF
compensation is usually performed using static VAr compensators (SVCs). Though, these devices
are effective in mitigating flicker whenever controlled properly, their performance is limited due to
inherent delays and their harmonics generation.
A control method for the thyristor controlled reactor/fixed capacitor (TCR/FC) was presented to
improve the dynamic response using a fast detection method [1]. This method is based on the
substitution of conventional filtering by an efficient, faster, and simpler process with a lower
computational burden and it guarantees that the compensation process will perform within 1=4 cycle ofthe fundamental period of the line voltage.
On the other hand, a solution of obtaining an optimum flicker reduction at a comparatively
low compensator power rating by an anti-windup extension of the controller is presented in
Reference [2]. The dimensioning of a SVC for a steel plant grid with several operating EAF,
taking the economic dimensioning of the compensator into account, is also demonstrated in
this paper.
orrespondence to: Arash Dehestani Kolagar, Department of Electrical Engineering, Faculty of Engineering,
Frequency characteristic of the filters impedance as well as the ones of the system impedance.
# 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
0 1 2 3 4 5-15
-10
-5
0
5
Time (Sec)
Qp
cc (
MV
Ar)
Reactive Power at PCC, with FC but not TCR
Figure 5. Simulated reactive power at 400 kV busbar after applying the harmonic filters.
L/2
SW1 SW2
L/2
Figure 6. The thyristor controlled reactor configuration.
REACTIVE POWER COMPENSATION IN A STEEL INDUSTRIAL PLANT 829
If the system impedance at a special harmonic is low, the corresponding filter to that harmonic does
not operate well and vice versa, i.e., if both the system and filter impedance have large values, it causes
very high harmonic voltages in the system. Ordinarily, a series resistor is inserted at each phase of the
harmonic filter whose value is obtained from the following equation [9]:
R ¼ Xn
Q(5)
0 1 2 3 4 5-5
0
5
10
Time (Sec)
Qp
cc (
MV
Ar)
Reactive Power at PCC, with TCR/FC
Figure 7. Simulated reactive power at 400 kV busbar after applying TCR/FC.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
0.2 0.21 0.22 0.23 0.24 0.25-300
-200
-100
0
100
200
300
Time (Sec)
TC
R V
olta
ge
(V)
an
d C
urr
en
t(kA
)
TCR Current (kA)
TCR Voltage(V)*30
Figure 8. TCR voltage and current from t¼ 0.2 second to t¼ 0.25 second.
0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28-4
-3
-2
-1
0
1
2
3
4
Time (Sec)
Voltages a
t P
CC
(kV
)
Before Compensation After Applying TCR/FC
0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28-4
-3
-2
-1
0
1
2
3
4
Time (Sec)
Voltages a
t P
CC
(kV
)
Before Compensation After Applying TCR/FC
Figure 9. Voltages at PCC from t¼ 0.2 second to t¼ 0.28 second.
0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.281.5
1.6
1.7
1.8
1.9
2
2.1
2.2
2.3
Time (Sec)
RM
S V
oltages a
t P
CC
(kV
)
Before Compensation
After Applying TCR/FC
Figure 10. RMS voltages at PCC from t¼ 0.2 second to t¼ 0.28 second.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
830 A. D. KOLAGAR ET AL.
Figure 11. IEC Flickermeter block Diagram [10].
REACTIVE POWER COMPENSATION IN A STEEL INDUSTRIAL PLANT 831
where
Xn ¼ffiffiffiffiffiffiffiffiffiffiffiXLXC
p¼
ffiffiffiffiL
C
r(6)
and Q is quality factor in the range of 30�Q� 100.
It is important to note that in order to prevent of resonance, if a filter is designed for eliminating
a special harmonic, lower order harmonic filters should be designed and installed at connection
point. Filters installing should also be accomplished according to the harmonic order from low
to high.
Figure 12. IFL curve (a) and its corresponding CPF curve (b) for 1st second before open-loopcompensation.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
832 A. D. KOLAGAR ET AL.
By considering the fundamental reactive power, which is some percentage more than the
measured one, 2nd, 3rd, 5th and 7th harmonic filters have been designed. Table I shows the
values of the filters elements, that are obtained through the explained process in previous
section.
Figure 4 shows the frequency characteristic of the filters’ impedance as well as the ones of the system
impedance.
Figure 5 shows the measured reactive power at 400 kV busbar (QPCC) after applying the
harmonic filters. As seen in this figure, the injected reactive power by harmonic filters provides the
required reactive power for load compensation. The remnant reactive power should be absorbed
by TCR. By comparing Figures 3 and 5, it can be seen that the total measured reactive power at
PCC busbar before and after applying harmonic filters are approximately 951 and �8MVAr,
respectively.
5. FINAL COMPENSATION BY APPLYING THYRISTOR CONTROLLED REACTOR
Figure 6 shows the scheme of a static compensator of the thyristor controlled reactor (TCR) type.
By increasing the thyristor firing angle, the magnitude of the component of the current reactor
is reduced. This is equivalent to the increase in the effective inductance, or reduction in the
reactive power absorbed by the reactor. However, it should be pointed out that the change in the
reactor current may take place only at discrete points of time. It means that adjustments cannot
be made more frequently than once per half-cycle. Static compensators of the TCR type are
Figure 13. IFL curve (a) and its corresponding CPF curve (b) for 1st second after open-loopcompensation.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
REACTIVE POWER COMPENSATION IN A STEEL INDUSTRIAL PLANT 833
characterized by the ability to perform continuous control, maximum delay of one half cycle and
practically no transients. The principal disadvantages of this configuration are the generation of
low frequency harmonic current components, and higher losses when working in the inductive
region [10].
By installing the TCR at PCC busbar, remnant reactive power is absorbed by TCR. The TCR
operation causes harmonic creation. In fact, when the furnaces are in their operating stage, i.e., special
scraping and melting stages, different harmonics are generated and the thyristors of the TCR are turned
off because of the high reactive power consumption of the furnaces. Moreover, when the charge of the
furnaces is alleviated, specially in refining stage, the generated harmonics from furnaces are reduced
and TCR turns on in order to absorb the surplus reactive power. Therefore, some harmonics resulting
from TCR operation are generated. Consequently, existing harmonics should be eliminated by
harmonic filters.
After determining the reactive power, which should be absorbed by TCR, the TCR desired
susceptance (B) is obtained by [6]
B ¼ qTCR
V2(7)
where V is the root mean square (rms) voltage along the TCR.
Then, the conduction angle of the TCR thyristors (s) can be found from Equation (8) [6]:
B ¼ s � Sins
pXL
(8)
Figure 14. IFL curve (a) and its corresponding CPF curve (b) for 8th second before open-loopcompensation.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
834 A. D. KOLAGAR ET AL.
The firing angle (a) of the TCR thyristors can then be obtained from the following equation [6]:
aþ s
2¼ p (9)
It should be noted that, in practice, after obtaining TCR desired susceptance in each cycle, thyristors
conduction and firing angles are generally determined by means of an off-line computed look-up table.
Figure 7 shows the measured reactive power at 400 kV busbar after applying TCR/FC compensator.
As seen in this figure, the TCR can balance the reactive power at PCC busbar properly. Figure 8
demonstrates TCR voltage and current from t¼ 0.2 second to t¼ 0.25 second.
Figures 9 and 10 also display the voltages and the RMS voltages at PCC before and after
compensation from t¼ 0.2 second to t¼ 0.28 second, respectively.
6. APPLYING IEC FLICKERMETER BEFORE AND AFTER OPEN-LOOP COMPENSATION
The international electrotechnical commission (IEC) flicker meter simulates spectator reaction
independent of flicker source. In fact, as the flicker meter operates based on flicker distinction feeling,
measurement of flicker is more or less relative. The basis of flicker calculation by IEC flicker meter is
that flicker intensity in input waveform is considered according to the quantity proper to the flicker
intensity. In this manner, the value of the corresponding quantity shows the indication level of the
flicker at each moment in input signal and also is an indication of the intensity and weakness of the
flicker. The name of this quantity is IFL. The IFL variable is an indication of instantaneous flicker
Figure 15. IFL curve (a) and its corresponding CPF curve (b) for 8th second after open-loopcompensation.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
REACTIVE POWER COMPENSATION IN A STEEL INDUSTRIAL PLANT 835
value in each moment [11–15]. Figure 11 shows the block diagram of the IEC flicker meter that was
simulated for flicker measurements [11].
For obtaining the instantaneous flicker sensation curve or short-term flicker severity curve (Pst), IFL
curves were evaluated for 15-second intervals, so that Pst value in each second is obtained by
considering the IFL curve in the same instant. Finally, after having 15 points, instantaneous flicker
sensation curve is estimated. Figures 12 and 13 show the IFL curve and its corresponding CPF curve for
1st second before and after open-loop compensation, respectively.
Figures 14 and 15 show the IFL curve and its corresponding CPF curve for 8th second before
and after open-loop compensation, respectively. The IFL curve and its corresponding CPF curve
for 15th second before and after open-loop compensation are shown in Figures 16 and 17,
respectively, too.
Figure 18 shows the instantaneous flicker sensation curve for the PCC busbar voltage before
and after compensation. As it is seen in this figure, open-loop compensation method at some points
reduces flicker intensity and in several points deteriorates it. Therefore, as a result, it can be stated that
this controlling method of the TCR/FC cannot reduce the flicker intensity, effectively. In fact, TCR
absorbs average surplus reactive power injected by FC banks at each cycle and cannot follow the
reactive power variations. In each cycle, TCR absorbs the average reactive power measured at previous
cycle. Because of the approximate similarity of two successive cycles, the amounts of reactive power,
which should be absorbed by the TCR at these two cycles, are very close together. Therefore, it can be
stated that TCR cannot track down the random variations; however, it can approximately absorb
the average surplus reactive power.
The Pst values related to Figure 18 are demonstrated in the Table II from t¼ 1 second to t¼ 5 second.
Figure 16. IFL curve (a) and its corresponding CPF curve (b) for 15th second before open-loopcompensation.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
Figure 17. IFL curve (a) and its corresponding CPF curve (b) for 15th second after open-loopcompensation.
Figure 18. Instantaneous flicker sensation curves before (1) and after (2) open-loop compensation.
Table II. Pst values related to Figure 16 from t¼ 1 second to t¼ 5 second
Time t¼ 1 second t¼ 2 second t¼ 3 second t¼ 4 second t¼ 5 second
Pst value of curve (1) 3.9649 0.9200 0.2396 0.1293 0.0931Pst value of curve (2) 3.5375 0.9353 0.2437 0.1312 0.0946
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
836 A. D. KOLAGAR ET AL.
REACTIVE POWER COMPENSATION IN A STEEL INDUSTRIAL PLANT 837
7. CONCLUSION
In this paper, an industrial steel making plant, which was previously installed in Mobarakeh,
Isfahan, Iran, is considered for research. An open-loop controlled TCR/FC was applied for reactive
power compensation. Arc model in this paper is derived from sampling arc currents and their
corresponding voltages. Then, the modulated arc voltage is considered by a band-limited white
noise. This model can represent an actual arc V–I characteristic. The open-loop control scheme for
reactive power compensation is accomplished. Since the open-loop controller directly calculates
the requisite reactive power as a feedforward control, it gives a fast response to compensate the load
reactive power consumption. Simulation results indicate that although the consumed reactive power
is almost compensated, open-loop control of the TCR/FC can not effectively reduce the flicker
intensity.
8. LIST OF ABBREVIATIONS
CPF C
Copyrigh
umulative Probability Function
EAF E
lectric Arc Furnace
FC F
ixed Capacitor
IEC I
nternational Electrotechnical Commission
IFL I
nstantaneous Flicker Level
PCC P
oint of Common Coupling
SVC S
tatic VAr Compensator
TCR T
hyristor Controlled Reactor
ACKNOWLEDGEMENTS
Special thanks are dedicated to the center of research and technology at University of Isfahan for their fullsupports. The present research work was originally related to a research project, which was successfully carried outand finalized by the authors. Special thanks go to the people of Mobarakeh Steel-Making Company for their fullsupport to us.
REFERENCES
1. Lee W, Kim T. Control of the thyristor-controlled reactor for reactive power compensation and load balancing. IEEE
Industrial Electronics and Applications Conference, May 2007; 201–206.
2. Prinz S, Pietzsch H, Stade D. Optimal control of static VAr compensators in power supply systems with electric arc
furnace. IEEE Power Electronics and Applications Conference, 2005; 1–10.
3. Marchesoni M, Mazzucchelli M, Monne A. Hybrid reactive power compensation system: advanced control strategy
development. IEEE Power Technology Conference, Vol. 4, June 2003; 1–8.
4. Chen CS, Chuang HJ, Hsu CT, Tseng SM. Mitigation of voltage fluctuation for an industrial customer with arc
furnace. IEEE Power Engineering Society, Vol. 3, July 2001; 1610–1615.
5. Varadan S, Makram EB, Girgis AA. A new time domain voltage source model for an arc furnace using EMTP. IEEE
Transactions on Power Delivery 1996; 11(3): 1685–1691.6. Miller TJ. Reactive Power Control in Electric Systems. John Willey & Sons: New York, 1982.
7. Makram EB, Subramaniam EV, Girgis AA, Catoe R. Harmonic filter design using actual recorded data. IEEE
Transactions on Industry Applications 1993; 29(6): 1176–1183.8. Sadeghi SHH, Der-Minassians A, Kouhsari SM. Power System Harmonics. Amirkabir University of Technology,
Tehran Polytechnic Press: Tehran, 2003.
9. Wakileh GJ. Power System Harmonics, Fundamentals, Analysis and Filter Design. Springer: Berlin Heidelberg,
2001.
10. Dixon J, Moran L, Rodriguez J, Domke R. Reactive power compensation technologies, state-of-the-art review.
Proceedings of IEEE 2005; 93: 2144–2164.11. Wiczynski G. Simple model of flickermeter signal chain for deformed modulating signals. IEEE Transactions on
Power Delivery 2008; 23(4): 1743–1748.
t # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep
838 A. D. KOLAGAR ET AL.
12. Wang CS, Chen LR, Yang WR. Analysis of equivalent 10-Hz voltage flicker measurement using IEC flickermeter
standard. IEEE International Conference on Industrial Technology, April 2008; 1–5.
13. Gallo D, Landi C, Langella R, Testa A. Implementation of a test system for advanced calibration and
performance analysis of flickermeters. IEEE Transaction on Instrumentation and measurement 2004; 53(4):1078–1085.
14. Hernandez A, Mayordomo JG, Asensi R, Beites LF. A new frequency domain approach for flicker evaluation of arc
furnace. IEEE Transactions on Power Delivery 2003; 18(2): 631–638.15. Bhattacharyya S, Cobben S, Myrzik J, Kling W. Flicker propagation and emission coordination study in a simulated
low voltage network. European Transactions on Electrical Power 2010; 20: 52–67.
Copyright # 2010 John Wiley & Sons, Ltd. Euro. Trans. Electr. Power 2011; 21:824–838DOI: 10.1002/etep