Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
1
High Temperature Superconducting Cable for Power
Transmission Applications
2
Contents
l Design possibilities of superconducting power cables
w Warm and Cold dielectric concepts
w Benefits and drawbacks
w Grid operation of VLI (Very Low Impedance) cables
l Power cable development considerations
w Design drivers and goals
w Trade offs
l Projects
w Major projects for power cables
w Economical benefits
3
Warm Dielectric Cable Concept
Former Tube
High Voltage Dielectric
Inner Cryostat Wall
Outer Cryostat Wall
Copper Shield
HTS Tape
Outer Cable Sheath
Inner LN2 Coolant
Outer LN2 Coolant
Protection Layer
4
Cold Dielectric Cable Concept
Outer protective covering
Outer Cryostat Wall
Inner Cryostat Wall
Liquid Nitrogen Coolant
Copper Shield Stabilization
HTS Shield Tape
High Voltage Dielectric
HTS Tape
Core / Former
5
Cold Dielectric Alternatives
Three phases in one cryogenic envelope
Concentric phases / Triaxial
3 independent phases
6
Concentric (Triaxial) Concept
Former Tube
High Voltage Dielectric
Inner Cryostat Wall
Outer Cryostat Wall
Copper Shield
HTS Tape
Outer protective Covering
Inner LN2 Coolant
Outer LN2 Coolant
7
Comparison Cold and Warm Dielectric
cryogenic envelope
conductor
LN2cold dielectric
superconducting screen
cryogenicenvelope
conductor
LN2
concept 1:warm dielectric
concept 2:cold dielectric
warm dielectricouter magnetic field
8
Design Comparison
outer magn. field
AC-loss
transport current
dielectric
cryostat
HTS-tape amount
accessories
cable inductance
warm dielectric
yes
high
high
conventional
at high voltage
(difficult to access)
low
almost conventional
as conventional cables
cold dielectric
no
low
very high
new type (cold)
at ground potential
(easy access)
high (screen)
new design required
very low
coaxial
no
low
very high
new type (cold)
at ground potential
(easy access)
low
very difficult at high
voltage levels
very low
9
LIPA 138 kV HTS Cable Project – Phase 1
345 kV 138 kV
345 kVto ConEd
in NYC
East Garden City Sub-Station
Ruland RoadSub-Station
NewbridgeSub-Station
PilgrimSub-Station
138 kV138 kV138 kV
2005
9.1 km6 km 16.7km
DOE Co-Funded – 610m 138kV 2400A HTS Cable
345/138 138/69610 m
11
14
l Long Island Power Authority – east garden city substationl Electrical operating characteristics
w Operating voltage/current – 138kV/2400A ~ 574MVA
w Design fault current – 69,000A @ 15 line cycles (250ms)l Physical characteristics
w Installation – one 12” pipe (11,7”/ 297mm ID)
w Length – app. 610m
w HTS tape length – 128km
w Cold dielectric design with three individualcryogenic envelopes
l Hardware deliverablesw Three ~ 610m long phase conductors
w Six 161kV outdoor terminations
w One 10kW refrigeration system at one end
w Single joint to be tested in the lab
Worlds First Installation of a Transmission Voltage HTS Cable in the World
Cable design drivers - LIPA Project Fact Sheet
15
Cable Design Aspects
Various aspects of cable design need to be traded to reach the optimized design point during
l normal operation as well as
l fault current operation
Cable loss
Former diameter
Cooling stationdistance
Pipe size forinstallation
Pressure drop
Tape Ic
Cable corediameter
Fault current andduration
Optimized cabledesign
Cable rating
16
Technical Issue Cable Cooling
l Goal: cable temperature and pressure within specified range
l Trade off:
w Hydraulic diameter determined by
n Cable diameter
n Pipe size
w Distance between cooling stations (Length)
w Cable losses determined by
n HTS-tapes
n Cryostat
n terminations
Hydraulic diameter of the
cooling channel
Cable diameter
High Voltage
Fault conditions
Pipe size
Cable installation
Pumping port
Losses HTS tapes Cryostat
Termination
Length Distance between
substation
Cooling flow of liquid nitrogen 65 K < T < 77 K 3 bars < P < 20 bars
17
Cable Cooling Example LIPA Project
l Flow of LN2 in single phase and back in two phasesl Effect of second pump and / or cooling station analyzed
18
Technical Issue Fault Current Operation
l Goal: no bubble formation during fault event to avoid dielectric breakdown
l Trade off:
w Choice of materials
n Resistivity
n Thermal conductivity
n Thermal capacity
w Cable design
n Current sharing properties
n Heat exchange properties
Materials Resistivity Current
sharing Temperature
increase
Heat exchange
Cable design
Materials Thermal
properties
No bubble in the
dielectric during the
fault
19
Main HTS Cable Projects
Location Main Utilities Cable Use Status
partners Dielectric Number Characterictics
type of phases
U. S. A. Pirelli / ASC Warm 1 50 m / 115 kV / 2 kA Demonstrator Complete
Berlin Pirelli (ex-Siemens) [1] Cold 1 50 m / 110 kV / 2.1 kA Demonstrator Stopped
Italy Pirelli / ASC ENEL / Edison Cold 1 30 m / 132 kV / 3 kA Demonstrator ?
Detroit Pirelli / ASC Detroit Edison Warm 3 120 m / 24 kV / 2.4 kA Network Cryostat Issue
Paris Pirelli / ASC EDF Cold 1 50 m / 225 kV / 2.6 kA Demonstrator Complete
Tokyo SEI TEPCO Cold 30 m / 66 kV / 1 kA Demonstrator Complete
Tokyo SEI TEPCO Cold 3 100 m / 66 kV / 1 kA Demonstrator Complete
Albany (NY) SEI / IGC Niagara Mohawk Cold 3 350 m / 34.5 kV / 0.8 kA Network Ongoing
Japan Furukawa Cold 1 500 m / 77 kV / 1 kA Demonstrator Ongoing
Carrollton (Ga) Southwire / IGC Southern California Edison Cold 3 (rigid) 30 m / 12.5 kV / 1.25 kA Plant supply Complete
Copenhagen NKT / NST [1] Elkraft Warm 3 30 m / 36 kV / 2 kA Network Complete
Columbus (Ohio) Southwire American Electric Power Cold / (Tri) 3 300 m / 12.5 kV / 2.5 kA Network Ongoing
Kunming (China) Innopower / Innova [2] Yunnan Electric Power Warm 3 30 m / 35 kV / 2 kA Network Ongoing
Long Island (NY) ASC / Nexans Long Island Power Authority Cold 3 610 m / 138 kV / 2.4 kA Network Ongoing
[1] Cryogenic envelope supplied by Nexans. [2] Cryogenic envelope and dielectric supplied by Nexans.
20
Economical Benefits
l Shorter lengths of insertion – no tying back to existing EHV backbone system (VLI)
l Lower voltage – benefit through cheaper auxiliary equipmentl Greater controlability with use of Phase Angle Regulator – control of
power flow (VLI)l Expanded generator siting options because of lower voltage drop (VLI)l Reduced electrical losses
Superconducting power cables (particularly VLI) do not have to be cost competitive on a stand alone basis. Economical benefits of the „grid
solution“ are of interest
21
Conclusion
l Different cable design concepts with their benefits and drawbacks are well understood
l Very low impedance characteristic of cold dielectric cables is of big interest to utilities
l Cable design tradeoffs driven by the specifications are to be managed
l Current ongoing projects enter the step from laboratory setups to grid installations
l The LIPA cable project is the first one on transmission voltage level so far
Material investigation
short cable models
Laboratory setups
Network demonstrators
Market entry
22
Thank you for your attention