Transcript Prezentace aplikace PowerPoint

CIGRE Working Group C4.307:
„Resonance and ferroresonance in power networks
and transformer energisation studies”
Chairman:
Lubomir Kocis
EGU HV
Laboratory, Czech Rep.
Co-chairman:
Manuel Martinez
EdF,
France
Except of WG for ferroresonance, creation of WG Transformer
energisation studies was proposed in 2008
It was decided to join both topics, that are closely related, to
one WG.
WG members:
Name:
Herivelto Bronzeado
Orla Burke
Bruno Caillault
Zia Emin
William Phang
David Jacobson
Nicola Chiesa
Terrence G. Martinich
J. A. Martinez Velasco
Stephan Pack
Eung-Bo Shim
François Zgainski
In total: 14 members
Company:
CHESF
ESB International
EDF DTG
National Grid
ESB International
Manitoba Hydro
SINTEF
BC Hydro
UPC
TU Graz
KEPRI
EDF DTG
The first meeting:
Prague, 26 - 27 May 2010, EGU HV Laboratory
Attended by (9 members ):
Orla Burke, Bruno Caillault, Zia Emin, William Phang, Manuel Martinez,
Nicola Chiesa, Lubomir Kocis, Terrence G. Martinich, François Zgainski
Introductory contributions were presented dealing mainly with
transformer energisation during „black starts“
Proposed structure of the future document (brochure)
was discussed in the meeting
It will consist of two parts:
I. Resonance and ferroresonance in power systems
Proposed by L. Kocis, March 2010
a) Resonance
1. Resonant circuits and their characteristics (F)
2. Resonances in networks- recorded cases (T,F, O)
3. Analyse of possible resonant circuits
- series compensation, shunt compenation, FACTS (F)
- double cicuit lines, very long lines, capacitor banks, cables
- application of long HVAC cables
- non-standard circuits configured during restoration of system
operation (start from dark) - network islands, acceptable scenarios
(T)
4. Resonant overvoltages in hv and uhv networks - summary of risks of
their appearance
5. Mittigation techniques (O,..)
b) Ferroresonance
1. Oscillating circuit with core saturation
- free oscillations, lossless, damped
2. Key parameters
- magnetizing curves
- stray capacitance
- series capacitance
3. Feroresonance - steady state
- with small losses, large losses
- stability limits
- ferroresonance 50 Hz and subharmonic ferroresonance
4. Transition between low mode and high mode
- criteria for transition (50 Hz, subharmonic)
5. Ferroresonance of VTs in effectively earthed systems (Z, O)
- influence of circuit capacitances (grading capacitors of CB)
- role of VT magnetizing curve
- criteria for VT ferroresonance
- starting manipulations (switchings)
6. Feroresonances (FR) of VTs in non-effectively earthed systems (T)
- FR in tertiary circuits of large power transformers
- FR in power generator circuits
7. FR of power transformers connected to double circuit line ( Z, T,)
8. Consequencies of FR
- overvoltage, overcurrent, degradation, failures
- 50 Hz vs subharmonic
- requirements for degree of suppression
9. Measures for suppression of ferroresonance (O,T)
- additional losses
- devices for FR suppression
- VT design and dimensioning
- circuit configurations
II. Transformer Energization Studies
Proposed by Manuel Martinez, June 2010
1. Overvoltages and undervoltages due to transformer energization
a. Overvoltage generation due to LF harmonic resonance in weak
networks (transformer energization through long lines or cables) [L, T, F,
M]
· System restoration : Transmission and Distribution Transformers & Auxiliary
Transformers of Nuclear Power Plants
· Offshore wind farm transformers linked to the shore network by long AC
cables.
· Examples: field recorded cases
b. Voltage drop in distribution networks at the energization of transformers of
distributed generation units1 [M, T, H]
· Examples: Field recorded cases
2. Computing the overvoltages or undervoltages by simulation
a. Component modeling
· Transformers (single phase/three-phase core, saturation, residual flux, etc.)
[N,M, H]
· Overhead-lines and cables [M]
· Network equivalents
· Generators (X’’d simple models versus complete Park’s models, voltage
regulation…)[M]
· Loads
b. Random initial conditions: range & discretization
· Circuit-breaker closing times (dispersion between CB phases, discretization) [F,
L]
· Residual fluxes (values, phase distribution patterns, decay with time,
discretization) [F, N]
c. Assessing the sensitivity of the computed overvoltages to the
modeling of the upstream network (because overvoltages are due to
resonance, they can be very sensitive to small differences in the modeling of
the upstream network (location of impedance poles, etc.))
3. Evaluating the effects of the overvoltages : quantification of the stress
· Transformer withstand capability to TOV with harmonics [M, L]
· Surge arresters withstand capability to TOV with harmonics [M, L]
4. Mitigation techniques
· Controlled switching, shunt reactances, closing resistors, surge arresters,
network
modification, local magnetization… [F, M, T, L]
· Domain of application, effectiveness…
5. Some case studies: Simulation results vs. field measurements [F, M, N]
1 For instance, in France DU/U must be less than 5 %.
Next meeting of WG C4.307:
CIGRE 2010 General Session, Paris,
Thursday 26 August 2010 from 14 to 18 h
Room 332 M Level 3.5
Ferroresonance in Czech Transmission Network and Power
Generation
Two types of ferroresonance time to time occured:
Type A (400 kV)- Ferroresonance of CB grading capacitors with
VTs
Type B (MV) - Ferroresonance of VTs in tertiary circuits of large
power transformers and in generator circuits
Conceptual approach
Type A

- Measuring of magnetizing curves of all used types of VTs by
impulse method

- Simulation methods

- Finding of circuits sensitive to FR - field measurements

- Combining of VTs and CBs immune to FR
Type B

- Design of VTs with low magnetisation for tertiary circuits of
large power transformers and for generator circuits
Measuring of magnetizing curve by free oscillations
Figure Circuit for measurment and evaluation of VT magnetizing curve
AI(t) = uVT(t) – Rw . i(t)
(1)
(t) = ui(t).dt + (0)
(2)
Figure Reconstruction of magnetizing curve from measuring of VT free
oscillations, (record of voltage and current, integrated flux and resulting
magnetizing curve
Comparison of magnetizing curves of three different types of
VT used in the network 400 kV
VT3
4500
4000
3500
VT2
3000
3000
VT1
2500
2500
2500
2000
2000
2000
1500
1500
1500
1000
1000
1000
500
500
500
0
0
0,2
0,4
0,6
0,8
1
0
0
0
0,2
0,4
0,6
0,8
1
0
0,2
0,4
0,6
0,8
1
C=1 nF
180
VT3
160
VT2
Frequency (Hz)
140
VT1
120
100
80
60
40
20
0
0
200
400
600
800
1000
1200
1400
1600
Voltage (kV)
Figure Frequency of free oscillations as a function of initial voltage
Uco for three types of VTs and capacities 1 nF
1600
1400
DC
AC(amp)
Corona losses (kW/km)
1200
1000
800
600
400
200
0
0
200
400
600
800
1000
1200
1400
Voltage (kV)
Figure Corona losses in kW/km vs voltage
Figure Sensitivity of FR to paralel capacity
TABLE Combination of CB and VTs - risk of ferroresonance 50 Hz
Cs = 800 pF
(HPL420, VSV420.1)
Cs = 500 pF
Cs = 200
pF (VVR,
3AQ2)
Cp (pF) =
500 1000 1500 2000 500 1000 1500 2000 500 1000
OTEF420
yes yes
yes
NKF400-65
yes yes yes yes * yes yes yes
yes
VT1 420
yes
yes
UTF 420
yes
VEOS 420
yes yes
yes yes yes
SVS
650 550
420/1G**
kV kV
SVAS420/1G -
In very short bays equipped by CBs with grading capacitors more than 400
pF, we install VTs with very high immunity to FR (high knee of
magnetizing curve, air gap in core)
Type B Ferroresonance in MV circuits of large power
transformers or circuits of generators
Measures for suppression of FR in systems with isolated neutral
Historically - damped resistor in opened delta secondary of VTs
Experience showed, that damped resistor is effectively able to suppressed
steady-state FR but, in some cases not its initial transition part.
Modern automatic systems of remote control of substatios ( e.g.action as a
response to earth fault) have very fast times of reactions with adjusted
time of insucceptibility of about 150 miliseconds.
That is reason, why FR in tertiary must be suppressed from the beginning
including its transition part lasting several period.
Advanced designs of digital relays with very low burden led to construction
of VTs with very low power of order of units of VA.
It enables much more effective manner how to suppress FR in MV circuits.
Classic VTs have steady state magnetisation about 0,8 T.
Todays designs of VTs can be constructed (keeping required accuracy 0,2%)
with magnetisation 0,4 - 0,6 T.
It was proved, that after replacing clasic VTs in circuit with ferroresonance by
VTs with low level of magnetization 0,4 T, FR was suppressed at all.
ACM - Automatic central monitoring - on line diagnostic
system
Fault recorders are sources of a huge amount of data that are not
always effectively used for equipment service condition
assessment.
They are usually triggered only from relay system operation or by
exceeding a threshold value of phase current or voltage (usually
120% Unr.m.s. and 150-200% In but it can be set as needed).
However there is nothing that prevents their triggering from
substation control systems to record normal service switching
operations too. Doing that the fault recorders become unique
sources of data describing some transient events in substations.
Fault recorders provide basically two types of information:

current and voltage curves (single phases and zero) with
sampling frequency 1 kHz for 0.2 to 0.3 seconds before and 3
to 5 seconds after the fault recorder function was triggered

timing characteristics (the beginning and the end) of different
signals, e.g. start and end of protection relays, start and end of
O or C impulse, transfer of the impulse to CB coils (O,C), start of
pole discrepancy, start of CB interlocking (for auto-reclosing and
for opening operations).
But with general triggering condition (fault, overvoltage,
overcurrent + before every switching), fault recorders can record
many other transients than only short circuit faults.
Data transfer to central network server
The records from fault recorders are transferred to terminals (one terminal
in every substation) and then once a day records are sent from
substation terminals via WAN (intranet) to central CEPS server in
Prague .
AROPO
AROPO is a modular system consisiting of independent software modules
- that have differrent functions - recognition of specific event or
evaluation of the records.
There are already running 12 AROPO expert modules called e.g.
Module QMPRU – can recognize a restrike during circuit breaker opening
Module PREP – can recognize and calculate levels and durations of
temporary overvoltages in substation bays
….
Module FERO: The module is able automatically to recognize transient
or stable ferroresonance 50 Hz or subharmonic in any record of
failure recorders collecting from the all substations of the CEPS
network.
Module FERO is in operation 6 months. It found two cases of steadystate subharmonic ferroresonance of VTs 16,6 Hz in bus breaker
circuits. This subharmonic ferroresonance is not dangerous for VTs.
The circuits were subjected to analysis of sensitivity to possible
appearance of dangerous ferroresonance 50 Hz.