You are on page 1of 10

WSEAS TRANSACTIONS on

INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

Study about the Process Control of an Electric Arc Furnace using


Simulations based on an Adaptive Algorithm
MANUELA PANOIU1, CAIUS PANOIU1, IOAN ORA2, ANCA IORDAN1
1
Faculty of Engineering Hunedoara,
2
Faculty of Electrical Engineering
Polytechnic University of Timisoara
1
Revolutiei street nr. 5, cod 331128
2
Vasile Parvan street, no. 2, Timisoara, cod 300223
ROMANIA
{c.panoiu, m.panoiu, anca.iordan}@fih.upt.ro, sora.ioan@et.upt.ro
Abstract: - The electric arc furnaces are a very large power load, determining the negative effects on the power
quality (harmonics currents, unbalanced load, and reactive power). For a maximum efficiency of the power
consumption it is necessary to use an automat system for control the harmonics filters, the reactive power
compensation installation and the electrodes position, in order to obtain a high value of power factor and a
maximum efficiency. In this paper is used an adaptive algorithm for process control for an Electric Arc
Furnace. The method is validating using simulation in PSCAD EMTDC software dedicated to Power Systems.

Key-Words: - Adaptive control, LMS algorithm, active power control


devices nor reactive power compensation. The
detailed results of these measurements were
presented in [7]. From the ones previously presented
resulted that the elements that contribute to the
development of an action concerning the
improvement of the electric powers quality can be
grouped in:
the capacitors fix battery in Y connection used
for compensation of the constant reactive power;
14
capacitors
battery
independently
connectable, in Y connection , used for
compensation of the variable reactive power;
an Adaptable Balancing Compensator (ABC)
achieved with 3 susceptances controlled by
thyristors in connection used for load
balancing as well as for the compensation of the
difference between the reactive power installed
in the capacitors batteries and harmonic filters
and the necessary of reactive power until the
obtaining of a unitary power factor;
4 filtration blocks of filter in Y connection used
for filtration of current harmonics 5,7,11 and 13.
Using of these installations aiming the
fulfillment of the functions for which they were
designed need the utilization of an intelligent system
of their control. The control system must allow the
online determination of the electric values of the
EAFs installation. The system must also calculate
the necessary step for compensating the reactive
power as well as the value of the susceptances

1 Introduction
In the last decade the EAF (Electric Arc
Furnaces) are very large used to the steel making
industry. But, the electrical installation of the
electric arc furnace is a massive generator of the
reactive power, harmonics currents and unbalanced
currents. These effects are very harmful for the
electric power supplying line and for the others
consumers. Because the inductive load character of
the electric arc furnace, the reactive power
component are significant, following to the
diminution of the active power factor and therefore
to the decreasing of the efficiency of the electric arc
furnace. In scope of improving the efficiency of the
entire installation it is necessary to use a complex
installation for reactive power compensation,
harmonics current filters and load balancing. These
installations must be connected to an automat
system for an efficient realtime control.

2 Designing the control system of the


EAF
The design of the control system was made
following the measurements made in an industrial
plant. The measurements were made at a 3-phase
power supply installation of a 3-phase EAF of 100 t,
to which were not connected the filters for the
current harmonics, neither the load symmetrisation

ISSN: 1790-0832

1628

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

The diagram of the control system proposed to


be used at the loads compensation balancing
filtration is depicted in fig. 1. In [7], [8] and [5] was
calculated the values of these elements.

necessary for load balancing. Based on these values,


having in view the chosen constructive solution of
adjustable susceptance with tyristors, is calculated
the control angle of each tyristor;

Ssc=1100 MVA (min.)

160 MVA
3 x TT

30 KV
3 x TC

ABC
Furnace
transformer

14 steps

FA5 FA7 FA11 FA13

ADAPTING BLOCK
(AB)

Impulse Generator Block


for Thiristros Command
(IGBTC)

Contactors Block
Command (CBC)

6 analog
signals
Data acquisition
board

Fig. 1.The system diagram for process control of the EAF


connecting like in figure 2. The outputs of these
meters permit to obtain the rms values.
The electrodes position controlling is performed
taking into account on the real condition existing on
the considered industrial plant. The maximum
motion speed of the electrodes is of 3 m/min (0.05
m/s) and is reached in emergency regime, its
variation being achieved as in fig. 2; The electric
arcs length can be modified from zero to a
maximum value determined by limiting the
integrators output, fig. 2; The calculus of the drop
voltage is made based on the electric arc model ([7],
[8]), the implementation diagram being also in fig.
2; The electric arcs length can be modified from
zero to a maximum value. Adjustment of the
electrodes position is made independently on each
phase. Simulation of the electric installations
operation modifying the electric arcs length was
made initially without harmonics filters, power
compensation or load balancing. It was considering
the electric arcs initial value l 0 = 16 cm as well as
the electrodes initial speed v1 = v2 = v3 = 0 m/min.
Then, was command the lowering-down of the

3 Simulation of the EAF functioning


For validating the proposed control system it was
made a simulation of the EAF using PSCAD
EMTDC simulation program [10]. For simulation it
was use an electric arc model, depending on the
nonlinearity of the electric arc. This model was
presented in [7] and [8]. Based on this model it was
made a simulation for the entire electrical
installation of the EAF and for the propose control
system. The PSCAD simulation scheme is depicted
in fig. 2.

3.1
The EAF functioning simulation on
controlling the electrodes position
The electrical items variation in different
functioning regimes can be done only if we consider
an arc length variation between 0, corresponding to
the short-circuit regime, and a maximum value. The
maximum value is determined in such a way that the
electric arc is burning. For observing the variation
area of the powers on the supplying line the active
power meters and reactive power meters was

ISSN: 1790-0832

1629

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

considered maximum length of the electric arc


lmax = 26 cm .

electrodes up to the fulfillment of the short-circuit


condition. After approximate 8 seconds it was
command, independently on each phase, the liftingup of the electrodes with different speeds up to the
The HV-MV
transformer

RL

AA

The MV-LV transformer


(Furnace transformer)

100 [MVA]

Um1
RL

BB

#1

.000001 [ohm]
Im2

.000001 [ohm]
Im3

#2

Um2
RL

CC

110 [kV]

C
30 [kV]

.000001 [ohm]
Im1

A A
B B

#1

#2

30kV

C C

Ul1 9.5422 E-6 [H]


IA2

Ul2 8.9416 e-6 [H] 3.64 e-3 [ohm]


IA3

0.565 [kV]

Um3

IA1

73.0 [MVA]

Ul3 9.5422 e-6 [H]


0.372e-3[ohm]

RA1

0.05

0.05

RA2

RA3

P
A
Power
Q
B
RMS

Electrodes speed control


v1 (m/s)
v2 (m/s)
v3 (m/s)

6.9 e-3 [ohm]

0.05

The three
phase electric
arc model

Measure active
power
RMS

-0.05

-0.05

-0.05

-0.001

-0.001

-0.011

A
D
0.0

Arc speed on
phase 1

v1

1
sT

v2

v3

l1

l2

2.0

l3

Uth1

F
A

Arc length on
phase 1

l1

-0.5

Measure
reactive power

B Comparator

l1

F
l0

TIME

v1

Uth1

The drop (threshold)


voltage on phase 1

Uth2

Uth3

Fig. 2. The PSCAD simulation scheme for measure active power and reactive power.
The calculus of the arc length based on the electrodes speed
power, considering the symmetrical short network,
has, according to the circles diagram, the value

In this way it was covered practically the entire


operation domain, from the short-circuit regime up
to the fulfillment of the conditions in which the
electric arc does not ignite anymore. The simulation
results are presented in fig. 3 and 4. One can
observe that the highest value of the active power is
obtained when the value of the arc length is
aprox.16 cm. The reactive power is positive
regardless the working regime, having values
between 15-100 MVAR, being therefore necessary
the utilization of the reactive powers compensation
installation. One can observe that the domain in
which the reactive power should be compensate is
higher than the one chosen in case of designing the
reactive powers compensation installation from [8].
This is due to the fact that the simulation included
also the short-circuit regime where the reactive

ISSN: 1790-0832

Qsc = 2 Scn = 103,23MVAR

(1)
In the electrodes short-circuit regime are obtained
maximum values of the currents on the three phases,
on the both supply lines and minimum values of the
voltages, fact due to the high loading of the 3-phase
transformers. The rms values of the currents and
voltages, fig. 3, are different between the phases
because the different values of the load impedance
and because of the different values of the arc lengths
on each phase. The different values of the load
impedances are obtained from the values of the
resistances and inductivities from relation (2) and
(3). These are the real values from the electrical

1630

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

Lr1 = Lr 3 = 9,5422 H ,

installation of an EAF in an industrial plant in


Romania.
Rr1 = 0,6908 m,

Rr 2 = 0,3640 m,

(3)

Lr 2 = 8,9416 H .

(2)

Rr 3 = 0,0372 m,

180

IA1

IA2

IA3

Uj1

Uj2

Uj3

Im1

Im2

Im3

Um1

Um2

Um3

160

Arc current (kA)

140
120
100
80
60
40
20
0

Voltage in sec. furnace transf

0.400
0.350
0.300
0.250
0.200
0.150
0.100
0.050
0.000

Current in primary furnace transf.

3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00

Voltage in primary furnace transf.

18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0.0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

Fig. 3 The rms values for currents and voltages in the secondary in primary voltage transformer

ISSN: 1790-0832

1631

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

Active power (MW)

50

25

Reactive power (MVAR)

150

125
100
75
50
25
0
0.400

Ust1

Ust2

Ust3

l1

l2

l3

Arc drop voltage (kV)

0.350
0.300
0.250
0.200
0.150
0.100
0.050
0.000
0.300

Arc length (m)

0.250
0.200

Maximum arc
length

0.150

Shortcircuit
regime

0.100
0.050

Electrodes speed

0.000
0.0350
0.0300
0.0250
0.0200
0.0150
0.0100
0.0050
0.0000
-0.0050
-0.0100
0.0

v1

v2

2.5

5.0

7.5

v3

10.0

12.5

15.0

17.5

20.0

Fig. 4. The variation of active and reactive power, drop voltage, arc length and electrodes speed

ISSN: 1790-0832

1632

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

was used the diagram presented in fig. 5. This


condition contains the 4 filters on the harmonics
5,7,11 and 13 and the reactive power compensation
installation composed by the constant part (in Y
connection) and the adjustable part in steps. The
values of the elements are the ones designed in [7]
and [8].

3.2 Simulation of the active power control


systems operation following the reactive
powers compensation and filtration of the
harmonic currents
To simulate the operation of the power control
system in different regimes using the reactive
powers filtration and compensation installation, it
The passive filters for 5, 7, 11 and 13
harmonics

23.04 [uF]

23.04 [uF]

23.04 [uF]

8.64 [uF]

8.64[uF]

8.64[uF]

5.76 [uF]

5.76uF]

5.76[uF]

5.76uF]

5.76uF]

5.76[uF]

0.01759 [H]

0.01759 [H]

0.01759 [H]

0.02393[H]

0.02393[H]

0.02393 [H]

0.01454[H]

0.01454 [H]

0.01454 [H]

0.01041[H]

0.01041[H]

0.01041H]

Furnace
transformer

Um2
.1 [ohm]
C
CIm3

Um3

#2

Uj1 0.00000934 [H] 6.9 e-3 [ohm]


B
Uj2 0.00000934H
C

30kV
0.6 [kV]

3.64 e-3 [ohm]

Uj3 0.00000934 [H]


0.372e-3[ohm]

RA1

The three
phase electric
arc model

P
RMS

#1

RA2

Power
Q
B

The three
phase variable
capacities

73.0 [MVA]

RA3

HV-MV
transformer

47.52 [uF]

30 [kV]

47.52 [uF]

110 [kV]

47.52 [uF]

#2

Um1
.1 [ohm]
B
BIm2

CC

#1

.1 [ohm]

RL

BB

Im1
A

100 [MVA]

+C

RL

AA

+C

RL

P P

RMS
A
D
0.0

B Comparator

+
-0.5

2.0

Fig. 5. The PSCAD simulation scheme for EAF with harmonics filters and power compensation installation
The PSCAD simulation scheme for choosing
compensation step is depicted in fig. 6.
In [10] was presented the powers dependency by the
drop voltage. Thus, in [10] was show that active
power reaches up to a maximum value for a certain
value of the drop voltage. Therefore, the active
power dependency by the drop voltage is a
monotone increasing function for drop voltage
values between 0 and a value corresponding to
maximum active power. For these values the
dependency of active power/drop voltage is a
bijective function. Because the drop voltage
depends linearly by the electric arcs length, it
results that also the active power depends on the
electric arcs length. Based on these remarks, the
active powers iterative adjustment algorithm
proposed by the de authors is based on the

To ensure the reactive powers compensation on the


entire duration of the active powers control process
it is necessary that, depending on the reactive
powers momentary value, to connect or disconnect
one compensation step at a time.
Choosing of the compensation step is made as
follow:
- If the reactive power is situated within the
range 4,00 4,00 MVAR
the
compensation step does not modify;
- If the reactive power is higher than 4,00
MVAR a new compensation step will be
introduced;
- If the reactive power is lower than - 4,00
MVAR a compensation step will be
disconnected.

ISSN: 1790-0832

1633

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

modification of the electric arcs length depending


on the active power desired to be obtained.
Assuming that at iteration n the arcs length is l (n ) ,
and the active power is P(n ) , the arcs length at
iteration n + 1 will be given by the relation
l (n + 1) = l (n ) + e(n )
(4)
where
e(n ) = P0 P(n )
(5)
are the error by which is obtained the imposed
active power P0 at iteration n, and represents an
adapting factor. It is obvious that if the value of the
active power obtained at iteration n is higher than
the value of the imposed active power P0
is
.

necessary to reduce the electric arcs length and


opposite, fact ensured by the presented algorithm.
[9], [11]. This algorithm is known as the LMS
algorithm (Least Mean Square) or the stochastic
gradients algorithm, being, due to its simplicity, the
most used algorithm implemented in the current
systems. Choosing of the adapting factors values is
made taking into account its influence upon the
algorithms main characteristics: the algorithms
convergence speed and the adjustment error.
Were obtained the results presented in fig. 7 for an
adapting factors value = 0,000001 and in figure 8
for an adapting factors value =0,000005.

Determination of direction for reactive power compensation


Monostable

D
Q

1
sT

B Comparator

4.17

Clear
-4.17

Monostable

B Comparator

-1.0

The calculus of equivalent capacity

14.4

Cstep

N
C

D + +
F
C

N/D
D

Cech
3.0

Fig. 6. The PSCAD simulation scheme for choosing compensation step. The calculus for equivalent capacity

4 Conclusion
obtained around the desired value is higher, the
algorithm being possible to lose the convergence.
Smaller values of the adapting factor allow the
obtaining of a smaller dispersion of the system
outputs values but the convergence speed is
smaller.

By using harmonics filters, load balancing and


reactive power compensation the functioning regime
of the UHP EAF can be improve by controlling the
active power. For higher values of the adapting
factor allow the obtaining of higher convergence
speed of the control algorithm, but the dispersion

ISSN: 1790-0832

1634

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Active power (MW)

50

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

P0

40
30
20
10

The arc drop voltage (kV)

The arc length (m)

Equivalent capacity (micro F)

Reactive power (MVAR)

0
100
90
80
70
60
50
40
30
20
10
0
120

100
80
60
40
20
0
0.160
0.140
0.120
0.100
0.080
0.060
0.040
0.020
0.000
0.200
0.180
0.160
0.140
0.120
0.100
0.080
0.060
0.040

l1

l2

l3

Ust1

Ust2

Ust3

0.0

5.0

10.0

15.0

20.0

Fig. 7 Variation of active power (impose and simulated), reactive power, equivalent capacity, arc lengths and
drop voltages for =0.000001

ISSN: 1790-0832

1635

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Active power (MW)

50

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

P0

40
30
20
10

The arc drop voltage (kV)

The arc length (m)

Equivalent capacity (micro F)

Reactive power (MVAR)

0
90
80
70
60
50
40
30
20
10
0
-10
-20
225
200
175
150
125
100
75
50
25
0
0.225
0.200
0.175
0.150
0.125
0.100
0.075
0.050
0.025
0.000
0.250
0.225
0.200
0.175
0.150
0.125
0.100
0.075
0.050
0.025

l1

l2

l3

Ust1

Ust2

Ust3

0.0

5.0

10.0

15.0

20.0

Fig. 7 Variation of active power (impose and simulated), reactive power, equivalent capacity, arc lengths and
drop voltages for =0.000005

ISSN: 1790-0832

1636

Issue 11, Volume 5, November 2008

WSEAS TRANSACTIONS on
INFORMATION SCIENCE and APPLICATIONS

Manuela Panoiu, Caius Panoiu, Ioan Sora, Anca Iordan

References:
[1] IEEE recommended practice for monitoring
electric power quality. Standard IEEE std.
1159-1995
[2] Montanari, G.C., Loggini, M., Cavallini, A.,
Pitti, L., Zaminelli, D. (1994), Arc-Furnace
model for the Study of Flicker Compensation
in Electrical Networks, IEEE Transactions on
Power Delivery, vol. 9, No. 4, pg. 2026-2036.
[3] Tang, L., Kolluri, S., Mark, F. Mc-Granaghan,
Voltage
Flicker
Prediction
for
two
simultaneously operated Arc Furnaces, IEEE
Trans. on Power Delivery, vol. 12, No. 2, 1997.
[4] Panoiu M, Panoiu C, Modeling and simulating
the AC electric arc using PSCAD EMTDC,
Proceedings of the 5th WSEAS Int. Conf. on
System Science and Simulation in Engineering,
Tenerife, Spain, Dec. 16-18, 2006
[5] Panoiu M., Panoiu C., Osaci M, Muscalagiu I.,
Simulation Result about harmonics filtering for
Improving the Functioning Regime of the UHP
EAF, Proceedings of the 7th WSEAS Int. Conf. on
Signal Processing, Computational Geometry and
Artificial Vision ,Vouliagmeni Beach, Athens,
Greece, Aug. 24-26, 2007, pg. 71-76
[6] Panoiu M., Panoiu C., Osaci M, Muscalagiu I.,
Simulation Results for Modeling the AC Electric
Arc as Nonlinear Element using PSCAD

ISSN: 1790-0832

[7]

[8]

[9]
[10]

[11]

[12]

1637

EMTDC, WSEAS Transaction on circuits and


systems, pp 149-156. vol 6, 2007
Panoiu M., Panoiu C., Osaci M, Muscalagiu I.,
Simulation Result about Harmonics Filtering
using Measurement of Some Electrical Items in
Electrical Installation on UHP EAF, WSEAS
Transaction on circuits and systems, vol 7, Jan
2008, pp 22-31.
Panoiu M., Panoiu C., Sora I, Iordan A., Rob R.,
Using Simulation for study the Possibility of
Canceling Load Unbalance of non-sinusoidal
WSEAS
High Power three-phase Loads,
TRANSACTIONS on SYSTEMS, Issue 7, Volume
7, July 2008, pp 699-710, ISSN: 1109-2777
Alexander, S. T., Adaptive Signal Processing,
Springer Verlag New York Inc., 1986.
Panoiu M, Panoiu C, Sora I, Osaci M, About the
possibility of power controlling in the ThreePhase Electric Arc Furnaces using PSCAD
EMTDC simulation program, Advances in
Electrical and Computer Engineering , vol. 7,
number 1 (27), 2007, ISSN 1582-7445, pp 38-43
Yuu-Seng Lau; Hussian, Z.M.; Harris, R., A
time-dependent LMS algorithm for adaptive
filtering, WSEAS Transactions on Circuits and
Systems, v 3, n 1, Jan. 2004, p 35-42
www.pscad.com

Issue 11, Volume 5, November 2008

You might also like