 
Dynamic Simulation of Arc Furnace Operation
and its Impact on Electrical Supply Network 
Electrical Supply Network
1 Introduction
During scrap melting in an arc furnace the arcs are subject to
disturbances, mainly
due to scrap cave in and the electromagnetic forces created by the
electrode
currents in the opposite electrodes. In addition to this the arc voltage
is not sinusoidal
and a source for generation of harmonics.
Due to these conditions the arc impedance and thus the electrode current
will
fluctuate, which can cause disturbances both in the low frequency range
(flicker) and
in the high frequency range (harmonics).
Measurements on arc furnaces have shown that the time dependent “steady
state”
reactive power
Q(t) = × u$(t) × i$(t) × sin (t)
1
2
ϕ
changes from one half cycle to the next and that the sign of the time
derivative of the
reactive power also change randomly.
In order to evaluate the impact of an arc furnace operation on the
electrical supply
network a computer model has been developed. With this model it is
possible to
simulate disturbances generated in the arc furnace and to calculate the
resultant
voltage fluctuation on the electrical supply line bus also with
equipment for reduction
of voltage fluctuations like Static Var Compensator (SVC) connected to
the furnace
bus.
Measurements on several installations have shown that the model can be
considered
representative both with regards to the furnace as a source of
disturbances and the
potential reduction of flicker with a SVC system installed. 
2 Simulation of Arc Furnace
For the dynamic simulation of flicker generated by arc furnace operation
it is
sufficient to evaluate the influence from disturbances within the
frequency range 1 to
30 Hz.
In the model the RMS value of the arc impedance is represented by a
random square
wave function (2 to 5 Hz) and a superimposed random function with a
frequency
twice the network frequency with a mean value equal to zero and a
standard
deviation half of the value for the low frequency disturbance.
A random generator is used for generation of standard distributed
numbers for both
amplitudes and frequencies in the three phases.
The fluctuation function is mathematically described as:
F(t) = F1(t) + F2 (t)
F t
F N F t t
f
F N F
f
t t
f
M
I
F M F s
T
M
I
F M F
T
s
T
1
1
1
2
1
1
2
( ) 1
( ) ( , ); ( )
( ) ( , ); ( )
=
+ − × − <
×
+ − ×
×
< − <
⎧
⎨ ⎪⎪
⎩ ⎪⎪
σ
σ
F2 t NF 0 F
1
2
( ) = ( , ×σ )
f N
T F T
M
= ( , T )
1
σ
where
FM = Mean value of the fluctuation function
σF = Standard deviation of the fluctuation function
TM = Mean value of the fluctuation period time
σT = Standard deviation of the period time
ts = Start time for each square wave
NF = Standard distribution sequence for the fluctuation function
NT = Standard distribution sequence for the period time
FM, σF, TM, σT, NF and NT are specified as input data. 











