CC II RR EE DD 18th International Conference on E lectric ity Distribution Turin, 6-9 June 2005 [626664]

CC II RR EE DD 18th International Conference on E lectric ity Distribution Turin, 6-9 June 2005

FAULT CURRENT LI MITERS FOR TRANSMI SSION & DI STRI BUTI ON NETWORKS

Ram PARASHA R, Christian SASSE, Ro bin BANKS an d Leslie FA LKINGHA M
AREVA T&D Technology Centre, United Kingdom
[anonimizat], [anonimizat], robin.banks@a reva-td.com, leslie.falkingha [anonimizat]

INTRODUCTION

In the liberalised el ectricity market, with strong emphas is on
renewable ener gy as an i ncreasing part of the gener ation
mix, much of the geogr aphically distributed new gener ation
is more econom ically fed into local distribution net work
operators’ (DNO’s) net works rather than the primary
transmission grid. Continuing gr owth in pow er dem and, t he
use of intelligent/active netw orks and the addition of
distributed gener ation, ( DG) which allow s multi-dir ectional
power flo w, all req uire tig hter interco nnectio ns. All of these
tend to r aise the fault current level of exis ting netw orks.
Fault Cu rrent Limiters ( FCL) can help networks to limit this
rise, particularly of asymmetrical fault current when X/ R
ratios are also increased by l ocal addi tion of gener ators,
transformers or series induct ors. A suitably priced r eliable
FCL that coul d be i ntroduced i nto existing pow er grids
without exces sively cos tly netw ork analys is would be
welcomed by utilities if it minimised cos ts of upgr ading or
reinfor cing components to handle fault conditions while
allowing increased nor mal pow er transmission. The FCL
may als o become attractive at transmission level voltages ,
132 kV and hi gher.
Maximum expans ion of load and D G is expect ed to be in the
medium-voltage distribution net work. Al though FC Ls for this
market were avai lable over forty year s ago t hey ar e not
normally econom ically feasible and at present there is,
therefore, no established m arket for such equi pment to
penetr ate. If, how ever, the cos t of pur chase, ins tallation,
maintenance and l osses of this kind of equipment can be
offset b y savings over th e lifeti me of other installed equipment
then a significant market for a suitably priced and technically
acceptable dis tribution FC L may be quick to develop.

FCL TECHNOL OGIES

Utilities h ave currently many conventional options to protect
their eq uipment against faults. For distribution networks,
options include circu it break ers, high impedan ce tran sformers,
Is-limiters, fuses, increased line voltages and current diverters.
The most com monly used dev ices are ch eap an d cos t-
effectiv e fuses with a few Is-limiters. Fo r higher voltage and
currents in the distribution system, the fuse is n ot used
becau se of its heatin g and the des ire for reclos ure without
maintenance or auto-changing of the fuse .
At the tran smission level, options include circu it breakers,
current limiting reacto rs, network splittin g and sequential
switching. Increased network impedance and generator
impedan ces are als o used.
Netw ork splittin g is a sim ple solution, but involves additional costs for monitorin g and the added equ ipment and features
needed to resto re secu rity and quality of supply.
With new DG tech nologies and many energy storage devices
being connected into the network, the role of voltage source
converters (VSC ) is becom ing increas ingly importan t. VSC s
contribute much lower fault current than either sy nchronous
or as ynchronous machines. They increas e the generator cost
considerably when the prim e source is neither d.c. n or at h igh
frequency, and introduce harmonics. Thoughts are being
given to the introduction of HVDC tran smission with FACTS
devices, w hich together offer fault cu rrent limitatio n and
control of pow er an d VAR flow. Thyristor based short-circuit
current limiting devices (SSCL) are also being developed to
offer fault current limiting at tran smission and distribution
levels [1,2], an d with energy storage dev ices [3].
For transmission and distribution networks, supercon ducting
FCL and SSC L are n ew emerging tech nologies. T he
developments of supercon ducting FCL and of solid-state
current limiters and switches are both being fund ed by US &
European governments to addres s the problem of increas ing
fault currents on the nations’ electric g rids. Each approach
has unique adv antages and targ eted applic ations. Advantages
of the supercon ducting device include inherent passivity (no
active monitoring or control mechanisms, promising hi gher
reliab ility), electrical “ transparency” (virtually only auxiliary
losses) in steady -state operation , and an environmentally
friendly nature when used in tran smission-voltage
applicatio ns. On the other hand, the solid-state d evice uses
relativ ely common solid-state sw itching tech nology, permits
activ e control of fault currents to enable better co -ordination
with other protection, and can be u sed in transmission and
distribution.
Like any other new technology, growth of the market is
expected to ex perien ce four periods : early invention and
discovery; small market needs; major tech nology changes;
then emergence of the major market. At present, the
supercon ducting FCL is probabl y in the secon d phase of its
growth. Many other new technologies and materials are
evolving and can be con sidered to be in the first phase. The
Is-limiter could be placed in the secon d phase, bu t its future
market poten tial is doubtful becau se of its small bu t finite risk
of failure an d as sociated h ealth & safety concerns [4].
Superconducting FCL is still co stly to develop, install an d
monitor in any trial, b ut much private m oney and very
extensive government fund ing is being devoted to bringing
both bulk and coat ed substrate supercon ductors from lab or
pilot production to bulk manufacture and use in demonstration
FCL. In particular, developments in the US h ave the
ambitious aim of meetin g commercial d emand for FCL s by
2007, w hen it is expected to peak .
CIRE D2005

Session No 1

CC II RR EE DD 18th International Conference on E lectric ity Distribution Turin, 6-9 June 2005

SPECIFICATION OF AN FCL

Although FCLs are generally designed to limit the current
before the first cu rrent peak, it is n ecessary to ask “Must the
current always be limited before its f irst peak?” Allowing
fault current limiting to occur after th e first cu rrent peak may
reduce th e cos t of some FCLs and may not be critical in
circu mstances where th e FC L has been installed to lim it the
fault current of switchgear w ith adequate peak withstand but
lower fault current breaking capacity. It has been sugge sted
that the first peak may not be crucial for applicatio ns in which
the FCL is in a bus tie [5 ]. The allo wable delay depends on
what heatin g and mechanical f orces the break ers, infeedin g
equipment and connectio ns to the fault can withstand without
damage. There is a pot ential opport unity and justification for
apply ing ‘slower’ FCL tech nology where all equ ipment can
be certif ied for the higher first peak but there is a much wider
market for full lim iting of all fault current, particularly if this
is lim ited to a m aximum value that allows manufacture of
larger batch es of a standard des ign of low cost switchgear,
e.g. vacuum breakers of not over 20 kA r.m.s. rating at 11 kV.
As a res ult of DG con nection s the fault current seen by an
individual protectiv e relay may increas e or decrease. An FCL
further modifies the protectiv e requ irements and must be
protected in event of an internal fault. A nother design issue is
what maximum short-term inrush current, which may be well
over 2.5 times th e rated load current [6], should be allowed
without degrading Power Quality by added voltage drops.
It is now a general view that an ideal FC L should meet th e
following basic req uirements:
• Zero impedance througho ut normal operation and large
impedance under fault conditions. (In superconducting
FCL a.c. lo sses w ill be small, an d those of other FCL
should be u nder 1% of rating).
• Provide rap id detectio n and initiatio n of limiting actio n
within 5-10 milliseco nds of onset of short circu it and limit
fault current to well below its p rospectiv e peak value.
• Be self triggered with fast reco very to permit auto-
reclo sure of transmission circu its on a persisten t fault.
Sugge sted capability is at least two faults, preferably more,
within a peri od of 3 – 15 s econ ds. If possible, the FCL
should handle several reclosing cycles.
• Be able to withstand the lim ited short-circu it current
for 50- 100 m s until protection and circu it break ers clear th e
fault (longer for back up protection).
• Must not be af fected by pow er switching in the
network and should cause negligible voltage sags, over
voltages or harmonics.
• Should protect ag ainst faults in a fail-safe manner, with
no risk to operating personnel, and must have a high level
of reliab ility with a minimum of maintenance. A
maintenance in terval of three y ears and a u seful operatin g
life span of 30 y ears have been propos ed.
• Should be com pact, lig htweight, cost-effectiv e and
environmentally friendly.
• Should be capabl e of performing the requ ired number
of operations througho ut its operational life. CIGRE survey sugge sts maximum operations required are < 5 while
EPRI survey shows customers preferrin g 50 or ev en more.
• Offer, if econ omical, a w ider m arket in marine d.c.

MARKET

Quantitativ e market information to assess th e potential o f
FCLs is difficult to obtain , as a tim e con suming and long-term
exercise. Cur rent w ork is b ased on discussio ns with
repres entatives in utilities , DNOs , government bodies , and
published literatu re.
On the ren ewable generation such as w ind power, to tal
forecas t future capacity gives the number of wind turbines
required and that will, in turn, set a lim it to the number of
FCLs necessary in the local n etwork.
A larg e percentage of new generation will be embedded
within the 11 kV network [7]. T he ratin g and capacity of
such 11 kV networks is thus likely to be ex ceeded, an d so
major reinforcement or upgrading 11 kV systems to a higher
voltage or altern atively addition of FCL s will have to be
considered. Typical generation sources considered are 1 MW
for hydro and landfill sources an d 5 – 10 MW for wind farm
schemes. Larger generation generally necessitates a higher
voltage connection. It is expected th at windfarms of capacity
10 – 100 MW will be con nected to 132 k V tran smission
network. Considering conventional distribution equipment &
installatio n practices, th e size of generators connected to an
11 kV network is cu rrently limited to 5 MW in total for
voltage control and to limit fault current [3]. At 132 kV level,
there is a significant market poten tial bu t the present market is
focussing on replacin g break ers in the abs ence of practicable
FCLs.
The FCL market can be split into two categ ories:
• High voltage (HV) tran smission:
52 kV ≤ Ur ≤ 145 k V (and abov e)
• Medium voltage (MV) distribution:
1 kV ≤ Ur ≤ 36 (40.5) k V
The medium-voltage distribution market could further be s ub-
divided i nto two cat egories:
• Voltage rating 24 / 36 kV, current ≥ 2000A
• Voltage rating 3.3 / 6.6 / 11 k V, current ≤ 1600A
Within the 36 kV segment, the majority of applicatio ns will
lie in the 11 kV distribution networks [8].

Global Demand

Global electricity generation is expected to ex perien ce
significant growth. New ly installed generation capacity of
3055 GW is expected by 2020 [9]. G rowth in Asia alone is
expected to be > 1000 G W. Global DG installed capaci ty
(generators < 10 MW) i s estimated to grow by 185 G W by
2020, bu t remains only 3% of new installed capaci ty in 2005,
6% in 2006 an d a s mall share ev en in 2020. Considering
added capacity of 185 GW, an d one FCL for each 5 MW of
new capacity , the maximum credible g lobal dem and for FCL
CIRE D2005

Session No 1

CC II RR EE DD 18th International Conference on E lectric ity Distribution Turin, 6-9 June 2005

works out to be at 37,000 s ites up to year 2020. A llowing for
inverter-based solutions at 4 0 – 50% to be introduced, it
leaves a m aximum requ irement of 22,200 sites for FCL or
other reinforcem ents of distribution systems by 2020. T he
FCL will ach ieve a larg e part of this market only if its cost
becomes low enough t o be attractive to developing countries.
In UK, the government target of 10% contribution by
renewables by 2010 an d 20% by 2020 s ugge sts a good m arket
for DG [10]. Total DG capaci ty today is abou t 8,500 MW.
10% renewables by 2010 requ ires abou t 8,000 MW of new
connections and 10 GW of CHP requires a further 5,500 MW.
Assuming 10,000 MW of connections at 11 k V level, and
need to limit fault lev el once per 5 MW added, it w ould
require 2000 lim itation s of fault lev el. How ever, DNOs [11]
have more con servative views. They see two poten tial
applications for FCLs; one for 132 k V transmission networks
and the ot her for 11 k V level distribution. The first
applicatio n is lik ely to arise f rom the big wind farms and DG
at 132 k V level. The secon d application will aris e from small
DG including CHP plants in existing 11 kV networks. In the
DNO’ s view, capacity at 132 k V level is likely to be added at
10 – 100 s ites and at 11 k V level (including some at 33 k V
level) at 100 – 1000 s ites. Considering switchgear u pgrade,
solid state inverter or frequency converter based solutions and
absence of an effectiv e FC L for new networks, it is wise to
consider opportu nities for FCL to be a m aximum of 40 % of
the total sites at 132 k V and 11/ 33 kV.
A Danish sur vey [6] has sho wn tha t the most obv ious need for
a FCL is in the 150/ 132 k V transmission networks and that
the applicatio n potential in the present network is rather
limited.
In US, it is estimated that 1300 t o 1900 n ew pow er plants are
needed [1] over the next 20 years. The DG m arket is likely to
be in the 1 MW to 50 MW size an d to be located near to the
customer sites. These w ill give rise to considerably higher
asymmetrical fault currents. For FC Ls rated at a n ominal
power of 10 – 20 MW, an annua l US market volume of
~ 1 billion US-$ has been estim ated [12].
Today the two biggest markets for pow er production are India
(with 7 – 8% growth) and China (with 10 – 12% growth).
However, these markets are governed by huge subsidies [13],
which may be a disincentive to the introduction of FCL . The
electrical in frastructure in most dev elopin g cou ntries is
relatively young and therefore no current market exists for
replacin g ageing equ ipment [14]. Gen eration capacity is
expected to grow from 1050 G W in 2000 t o 2400 G W in
2020. C hina repres ents roughly one-third of the aggregated
generation in the developing countries. T otal electricity
generation in China is expected to reach in excess of 3,700
TWh. All this will b ring significant ad dition to the
transmission infrastru cture an d in terco nnectio n an d
consequent increas e in fault current. Any capacity expansion
in the short-term is expected to be met with new sub-statio ns
with upgraded switchgear w here th ere w ill b e limited
potential for FCL s.
Table 1 s ummarises the poten tial m arket. Worl d
Mark ets Inverter
based
solutions1Other
FCL
solutions Total FCL s
required by
2020
UK
DG f orecas t 860 1,290 2,1502
DNO f orecas t 440 660 1,1002
Worl d 14,800 22,200 37,000
1Inverter b ased solutions calcu lated @40% of the total FCL
by 2010. 2Required by 2010 i f FCL available.
TABLE 1 – The Potential Market

PRICE

There are clearly two market segments for the FC L: the
distribution network and the tran smission network. As
discussed above, majority of distribution segment market will
be at the 11 kV operating voltage level and will m ainly be
governed by competitiv e prices. T he distribution market
demands an FCL that can com pete w ith existing switchgear.
Current price of switchgear varies between €7,500 to €25,000
depen ding upon the application . Therefore a majority of
customers w ill favour a FCL at a p rice not over 3 times that of
a break er. T his estimate ag rees with the survey carried out by
CIGRE [15]. T hus the guide pri ce of a suitable FCL for a
sizeable market penetration will lie betw een €22,500 an d
€75,000. For this market, recl osing time and number of
reclo sing operations will b e more im portant than in
transmission systems and that may push the price up.

In the tran smission network, a 145 k V series reactor m ay cost
between €1-3 million or even more depending upon the
current ratin g [16]. F urther, break ers cos t varies from
$55,000 f or a 145 k V, 3 k A break er to $200,000 for a
245 kV, 5 kA continuous and 80 k A interrupting capaci ty
break er. Therefore, it s eems logical th at an acceptable price
for a FCL in the tran smission network will vary between
€165,000- €3,000,000.
Another niche market ap plicatio n is in coupling FCL s with
generators – limiting the fault currents that might be sustained
by break ers and generators . These generator break ers are
very expensive [5]. Having a maximum voltage of 30 kV,
though m ost often use d at 13.5 kV and 22.9 kV, the price of
generator breakers varies over a range as sugge sted below:
(1) $400,000 f or 8 kA continuous and 80 k A interrupting
(2) $700,000 f or 13 k A continuous and 130 kA interrupting
(3) $1,500,000 for 26 k A con tinuous and 200 k A
interru pting.
Therefore, it is v ery attractiv e to limit the current that such
break ers and generators have to sustain. In this segment, an
acceptable F CL price is likely to be betw een €400,000 to
€1,500,000.
These co sts ref lect th e price lev el which will make FCL a
viable propos ition and allow substantial pen etration of the
markets. How ever, exact cos t figures for the new practical
devices can not be g iven becau se of their early stage of
developm ent and associated uncertain ty in the cos t of
material.
CIRE D2005

Session No 1

CC II RR EE DD 18th International Conference on E lectric ity Distribution Turin, 6-9 June 2005

Table 2 s ummarises viable prices for FCL.
FCL Applicatio ns Estim ated
Mark et Price
Power distribution market –
Medium voltage €22,500 – €75,000
Power Generation–up to 30 k V €400,000 – €1,500,000
Transmission network ≥ 72 k V €165,000 – €3,000,000
TABLE 2 – Estimated m arketabl e prices for FCLs

DISCUSSION

Fault current limiters h ave the potential to substantially
improve system performance, reliab ility and safety in this era
of increas ing pow er dem ands and complex interaction s.
Extensive funding of HT S based FCL tech nology will
continue despite its difficulty and present high cost, m ainly
because it is hoped that it w ill quite so on offer benefits not
otherwise av ailab le, in itially for distribution networks but
particularly for tran smission. Solid-state b ased FCL is
cheaper to develop than supercon ductor bas ed FCL, offers
better co -ordination with other transmission system protection
and cou ld also control load cu rrents. It is most likely that the
growing market will see FCL designs tied up with a suitable
circu it break er an d sold as single equ ipment.
Limiting the first peak may not alw ays be crucial w hen FCL s
reduce fault current in circu it break ers and other equipment
specially certif ied for this duty. In the 11 kV market, a fully
effectiv e FCL could redu ce ty pical cos ts of vacuum
switchgear if fault cu rrent is lim ited to a value such as to
allow standard low cost vacuum circu it break ers of some
20 kA rating to be employed througho ut most installations.
Introducing low cost su perconducting FCL has substantial
potential benefits th at have led to substantial in dustrial an d
government funding of research , development, and quantity
manufacture of supercon ducting materials followed by
proving of FCLs in the field. US f unding aim s at th eir
commercial introduction as early as 2007. T his is an
ambitious target that shows the value set u pon such devices.

REFERENCES

[1] D. Richardson, 2003, D evelopment and operat ional
advantages of a solid-state circu it break er with current
limiting, IEEE Bi annual Meeting, Pow ell Pow er
Electronic Company.

[2] D. Retzmann, 2004, U se of high-power thyristor
technology for short-circu it current limitatio n in high
voltage systems, IEEE W orkshop, Ber lin, Presentation
by Si emens.

[3] M. Mac D onald, 2004, Innovat ion i n electricity
distribution net works, OFGEM report 212281.
[4] Developm ent of a safety case for the use of current
limiting devices to manage short circu it cu rrents on
electrical distribution networks, 2004, D TI Final report
ref. URN 04/1066.

[5] A. M Wol sky, 2004, O perating Agent for IEA HTS Task
and Argonne National Laborat ory, Argonne, IL 60439,
USA – Private communication .

[6] J. N. Nielsen, J. J. O stergaard, 2001, Applications of
HTS fault current limiters in the Danish utility network,
CIRED 2001, 91.

[7] A. J. Beddoes , A. Collinson, 2001, L ikely changes to
network designs as a res ult of significant embedded
generation, DTI document ET SU/K/EL/0023/ REP,
produced by EA T echnol ogy.

[8] G. Strbac, N. Hatziargyriou, 2004, Mi crogrids – a
possible future en ergy configuration ?, IEA Sem inar on
Distributed Generation, K ey issues, Challenges, Roles,
Paris, France.

[9] A. Bauen, A. Hawkes, 2004, Decen tralis ed generation –
Technologies and market pers pectiv es, Presentation at
IEA Par is, France.

[10] M. Crouch, J. S cott, 2004, R ewiring distribution
regulation, OFGEM RAB Pr esentation .

[11] C. Mortley , 2004, EDF Energy, DNO R epresentative –
private com munication.

[12] R. Hott, 2004, A pplication fields of high temperature
supercon ductors, High-Temperature Super conduct ivity–
Engi neering Appl ications, Springer, Berlin , Germ any,
35.

[13] K. Kozloff, 2000, R enewable En ergy Strategies in
Developing a nd EIT Countries Under Restructured
Electricity Mark ets, Presented at “Acceler ating Gr id-
Based Renew able Ener gy Pow er Generation for a Clean
Environment”, Washington D C, USA.

[14] Electric pow er systems, 2003, CIGRE Jo int advisory
group SC 15/D1-JAG 02 T C, Repor t no. 225.

[15] H. Sch mitt, 2003, Fau lt current limiters – Report on the
activ ities o f CIGRE W G A3.10, Presentation by
Siem ens.
Fault current limiters in electrical m edium and high
voltage systems, 2003, CIGRE W orking G roup A3.10,
Repor t no. 239 .

[16] Fault current management plan, 2001, NYISO Joint TPAS
/ IITF Meeting, Pr esented by C on Edi son.
CIRE D2005

Session No 1