Superconducting fault current limiters
Fault
Current Limiters are regarded as key components for modern power systems.
Among all possible limiting devices SCFCL offers ideal performances:
negligible impedance in normal conditions and passive (i.e. reliable) switch
to high impedance state in case of fault. A great research effort has been
carried out in recent years, showing the feasibility of both the two basic
concepts of SCFCLs, the “Resistive” and the “Inductive” type. However,
the interaction of these devices with power networks is still little
investigated. In order to evaluate how a
SCFCL can enhance the performances of a power system, a circuit model of the
device needs to be introduced in power system simulators. The more this model
well reproduces the features of the limiting behavior, the more the technical
and economical benefits are accurately estimated.
At the Applied Superconductivity Laboratory of the University of Bologna the following type of superconducting limiters are investigated:
1. Magnetic shield type superconducting fault current limiter
2.
Transformer type superconducting fault current limiter with adjustable trigger
current level
1.
Magnetic shield type superconducting fault current limiter
A
Magnetic shield type superconducting fault current limiter consists of a SC
ring and a copper coil inductively coupled via a ferromagnetic core. A
numerical model for determining an accurate equivalent circuit of this device
has been developed. The details of the model are described in the section
“Electrodynamics of superconducting bulks”. The equivalent circuit
can be introduced in power system simulator in order to
investigate
the technical and economical benefits brought by the device the system.
In
order to validate the proposed circuit model a basic prototype of
Magnetic Shield type HTSFCL has been designed and manufactured. A BiSCCO
2212 ring with 130 mm inner diameter, 113 mm Outer Diameter and
50 mm height was first chosen as SC element. The value of critical
current density in self field at 77 K, measured by the ring
manufacturer by means of the 1 μV/cm criterion, was 810 A/cm2.
The E – J characteristics of the ring is shown in figure 1.1. |
Figure 1.1: E – J characteristic of BiSCCO 2212 ring |
The
maximum magneto-motive force produced by the coil which can be shielded by
this SC ring is equal to its whole critical current, that is 2025 A. Therefore
the minimum value of the device
quenching current, i.e. the minimum value of current flowing in the coil which
brings the ring to enter the SC/Normal transition, corresponding to the
case of perfect magnetic coupling, can
be calculated as the ratio of the ring critical current and the number of
turns of the coil. A minimum value of 7 A was chosen for the quenching and the
inducting coil was made with 285 turns,
arranged in 15 layers of 19 turns each and realized by using a 2
× 5 mm2 copper tape. The
inner diameter of the coil was equal to 160 mm and the thickness and height
were equal to 43 and 104 mm respectively. For what concern the magnetic core,
a column with circular cross section of 63 mm diameter, made of laminated
grain oriented silicon steel, was chosen. With this open core
configuration the ratio between the limiting and the non–limiting state
impedance of the device is too small to be of practical interest, but is large
enough to affect current and voltage waveforms in a way detectable by the
instrumentation of the laboratory. Moreover the axial symmetry of the whole
device is maintained, giving considerable time saving in the numerical
calculations. The height of the column,
chosen in order to have a value of the impedance ratio equal to about
3, was equal to 300 mm. Figure 2.1 shows the magnetic flux density
distribution inside the device, calculated preliminarily in order to specify
the dimension of the ferromagnetic core. The calculated values of the non
limiting and limiting inductances are also reported.
|
|
Fig
2.1.a. field distribution inside the device in non limiting conditions, LNL = 12.39 mH |
Fig
2.1.b. field distribution inside the device in limiting conditions,
LL = 35.86 mH
|
The circuit used for testing the device is represented in figure 2. The FCL prototype is series connected to a protective resistance RP of 1 Ω and to a load resistance RL of 9 Ω. |
|
The
circuit is supplied by an autotransformer connected to the 220V/50Hz AC
network and is controlled by a means of a triac SW1. The load
resistance is parallel connected to a second triac SW 2 which
command the fault. The voltage Vs of the source, the voltage VFCL across the device and the current IFCL flowing in the circuit have been acquired by means of insulated transducer modules connected to a multi-channel data acquisition card, with a sampling rate of 750 samples per second.
|
|
|
Two
digital I/O channels of the card have been used to control the insulated
drivers of the triacs. The system was managed by means of the LabVIEW
software. During
the tests the whole device was immersed in a liquid nitrogen bath.
|
The
equivalent circuit of the FCL prototype, under the assumption of linear iron
core, is shown in figure 3.1.
|
figure 3.1. equivalent circuit of the prototype |
Since
the numerical results obtained have shown that no appreciable current flows in
the radial branches of the equivalent circuit (this correspond to no
appreciable currents flowing in the axial direction in the actual device),
they have been neglected and the following reduced equivalent circuit of
figure 4.1 has been obtained.
|
figure 4.1. reduced equivalent circuit of the prototype |
Figure
5.1 shows the comparison between numerical and experimental waveforms of
current and voltage across the device during one of the short circuit test
performed.
|
figure 5.1: current and voltage across the device during a fault test |
As
it can been seen from the figure, as far as the normal operating (non fault)
condition is considered the numerical results are in a good agreement with the
experimental ones; this means that the equivalent circuit well reproduces the
shielding effect of the SC ring. Moreover, the current and voltage during a
short period after the fault occurrence are also well reproduced. A
discrepancy, probably due to the drop of critical current density caused by
the heating, begins to appear after some cycles.
[1]
F. Negrini, P.L. Ribani, A. Morandi, T. Nitta, S. Oshima, "Experimental
Analysis and Circuit Model of an Inductive Type High Temperature
Superconducting Fault Current Limiter", International Journal of Modern Physics
[2]
M. Fabbri, A. Morandi, F. Negrini, P.L. Ribani, “Magnetic Shield Type
Fault Current Limiter Equivalent Circuit”, submitted for publication on IEEE
Transactions on Applied Superconductivity
[3]
M. Fabbri, A. Morandi, F. Negrini, P.L. Ribani, L. Trevisani
“Experimental Validation of Magnetic Shield Type Fault Current Limiter
Circuit Model”, presented at 6th European Conference on Applied
Superconductivity, Sorrento Napoli - Italy, 14th – 18th
September 2003
2.
Transformer type superconducting fault current limiter with adjustable trigger
current level
A
transformer type superconducting fault current limiter consists of a primary
SC winding series connected with the test circuit and magnetically coupled (without
an iron core) with a secondary SC one, which is short circuited. The primary
coil is made of superconductors only in order to reduce the thermal load of
the cryostat. Due to the short circuit of the secondary superconducting
winding, in normal operating conditions the device shows negligible impedance.
When the current of the primary coil exceeds the trigger value, the current
induced in the secondary coil exceeds a critical value and the device switches
to an high impedance state. The trigger current can be adjusted by changing
the magnetic coupling of the windings.
The
experimental testing of this device has been carried out at the “Nitta Laboratory”,
The
University
of Tokyo.
Figure 1.2 shows a scheme of the device tetsted and a picture of the two SC
coils.
|
|
Figura
1.2: FCL device
Several
tests under various operating conditions have been performed. The
effectiveness of the limiting action and the possibility of adjusting the
trigger current have been confirmed by the experiments. For the fault cases
studied, the recovery time, i.e. the time required to switch back to a low
impedance state after the fault is removed, has been found to be zero, at
least in the limits of the control and measurements apparatus time scale.
In
order to investigate the interaction of the FCL with power networks a circuit
model of the device has been defined. The equivalent circuit is made of two
coupled inductors with auto inductances L1 and L2
respectively and with mutual inductance M. The inductor L2
is connected to a non linear resistor R12, whose
voltage-current relation is defined by means of the power law v
= k sign(i) |i|N,
which takes account of the SC/Normal transition of the winding. The parameters
of the equivalent circuit have been determined by examining the experimental
waveforms of current and voltage across the device during a fault.
Figure
2.2 shows the comparison between experimental and numerical waveforms
calculated by means ft he equivalent circuit.
Figure
2.2. current and voltage waveforms across the device