Professional Documents
Culture Documents
Non-detection zone
duration longer than AND
Interconnected 30
system generation
profile
feeder Pload
Qload 20
~ } non-detection zone
10
LOM
G t1 t2 t3 t4 24
Time [h]
Pgen
Qgen Risk of undetected island
LOM performance requirements
stability/security
LOM should be stable under remote faults cleared by
the utility system as well as under system dynamic events.
feeder Pload
Qload
~
LOM
G
Pgen
Qgen
Current Practice in dedicated LOM protection
Intertripping
Best performance but cost is high and can become overcomplicated in
some parts of the system.
50
V [V]
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-50
-100
t [s]
VT1
VT2
Example Vector Shift Relay Operation
Vector Shift is used to protect an Embedded Generator.
Calculate the voltage angle change if the generator output
increased from 15MW to 25MW as a result of an LOM event. Before
and islanding the generator operated at unity power factor pf1=1.
Assume that at the time of islanding the generator terminal voltage was
VT1 = 1/0pu.
Generator: Sn = 30 MVA Vn = 33 kV
Xd = 0.23 pu Ra = 0.05 pu
Xd
Xd
X d
IL R
Ed E d E d VT
Italy
New directive 84/2012/R/EEL (8 March 2012) was issued by the Italian
Regulatory Authority for Electricity and Gas (Aeeg), with the aim of
integrate the CEI 0-16 (Reference technical rules for the connection of
active and passive consumers to the HV and MV electrical networks of
distribution) with a technical document from TERNA (the Italian
TSO) defining the system requirements of the DG.
f<49.7
OR
f<50.3
V1<thr.
Systematic LOM relay testing
ETR 139
Systematic LOM relay testing ETR 139
Main objectives
To establish the behaviour, under various scenarios, of different Loss
of Mains (LOM) protection relays that are typically applied to the
protection of distributed generation installations
Produce a matrix of recommended settings for LOM
Propose standard test scenarios for LOM relays
Case studies
Scenario 1: Generator of 30MVA connected
to 33kV network
Scenario 2: Generator of less than 5MVA
connected to a section of 11kV network
Type of studies
Genuine loss of mains events sensitivity
Cleared system faults (where there is no actual loss of mains) on
adjacent circuits stability
Generator types
Gas turbine with synchronous machine
Wind turbine driven DFIG generator
Induction machine driven generator
(11kV scenario only)
DC/AC inverter connected source
(11kV scenario only)
Systematic LOM relay testing ETR 139
Network model
0.9
0.8
ROCOF setting [Hz/s]
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-10 -5 -2.5 0 2.5 5 10
Generation - local load balance [%Sn]
Systematic LOM relay testing ETR 139
8
ROCOF setting [Hz/s]
0
f1a f1ab f1abc f2a f2ab f2abc f3a f3ab f3abc
Fault scenario
Systematic LOM relay testing ETR 139
Main findings
The LOM performance is affected primarily by the generation
technology the most challenging is a synchronous generator
Stability causes more problems than sensitivity
Improvement in stability can be achieved by providing additional
time delay which does not significantly compromise the sensitivity
Significant difference in
performance between
different relay manufac-
turers was noted, mainly
in terms of stability
Non-detection
zone duration AND
(iii)
longer than the
acceptable limit,
e.g. 5s
Example NVD Requirement probability calculation
(ENW 11kV data Circuit 1)
0.7
Pload
Assumptions 0.6
Qload
P[MW], Q[MVAR]
0.4
o 5km of 95mm2 (5MVA) 0.3
50.2
50.15
50.1
50.05
Frequency (Hz)
50
49.95
49.9
Strathclyde
49.85
49.8 Manchester
49.75 Imperial
49.7
49.65
01:48:00.00 01:49:26.40 01:50:52.80 01:52:19.20 01:53:45.60 01:55:12.00
GPS Time
PMU Frequency Record
PMU Frequency
50.1
London
Manchester
50.05 Glasgow
50
49.95
f [Hz]
49.9
49.85
49.8
49.75
49.7
512 513 514 515 516 517 518 519 520
t [s]
PMU Phase Angle Difference
40
35
angle [deg]
30
25
20
512 513 514 515 516 517 518 519 520
t [s]
PMU df/dt Record
1.5 Strathclyde
Manchester
1 Imperial
0.5
df/dt (Hz/s)
0
01:48:31.97 01:48:32.83 01:48:33.70 01:48:34.56 01:48:35.42 01:48:36.29
-0.5
-1
-1.5
-2
GPS Time
df/dt calculated from PMU frequency
as 0.5s average
1.5 Strathclyde
Manchester
1 Imperial
0.5
df/dt (Hz/s)
0
01:48:31.97 01:48:32.83 01:48:33.70 01:48:34.56 01:48:35.42 01:48:36.29
-0.5
-1
-1.5
-2
GPS Time
df/dt calculated from PMU frequency
as 0.5s average
0.05 Strathclyde
Manchester
Imperial
0
01:48:31.97 01:48:32.83 01:48:33.70 01:48:34.56 01:48:35.42 01:48:36.29
-0.05
df/dt (Hz/s)
-0.1
-0.15
-0.2
-0.25
GPS Time
ROCOF relay testing
Three COMTRADE waveform records have been synthesised on a
cycle by cycle basis from the available PMU frequency profiles
and subsequently injected into the MiCOM P341 relay.
21 bus equivalent
transmission network model
1 is proposed
2
4
3
7YS was used to obtain the
load and generation data in
each zone.
PMU data captured during a major event in the UK transmission system 28 September 2012
70 50.05
50 GL
60
49.95
MN
50 49.9
LN
49.85
40 49.8
30 49.75
49.7
20 GL-MN 49.65
GL-LN 49.6
10
49.55
0 49.5
14:03:22 14:03:23 14:03:25 14:03:26 14:03:28 14:03:29 14:03:31 03:22.003:22.603:23.203:23.903:24.503:25.203:25.803:26.4
No PSSs included
50.05
70
50
GL
60
49.95 MN
50 49.9 LN
49.85
40
49.8
30 49.75
20 49.7
GL-MN
49.65
10 GL-LN
49.6
0 49.55
0.0 1.4 2.9 4.3 5.8 7.2 8.6 10.111.512.914.415.817.318.7 0.0 0.7 1.4 2.1 2.9 3.6 4.3 5.0 5.8 6.5 7.2 7.9 8.6 9.4
66
Adding stabilisers
0.15
4
Eigenvectors and 0.1
3
19
each lightly damped 0 9 1
10
-0.05
68
PMU Data - 28 September
TW TD n
Local frequency is measured by the relay
n n 1 2 ( f n f nref )t DG
LOM
GPS stamped
frequency signal
Protection
n calculated angle
fn frequency measured locally
fnref reference frequency received from the satellite
t angle calculation time step
A tripping signal is sent when n exceeds the preset threshold.
Satellite channel latency can be easily compensated using GPS
A. Dyko, G.M. Burt, P.J. Moore, I.A. Glover, J.R. McDonald, Satellite Communication Based Loss-of-Mains Protection, 9th International
Conference on Developments in Power System Protection, Glasgow UK, vol. 1, pp. 687-692, March 2008.
Satellite based LOM protection
5
Glasgow PMU
4 record as a DG local
angle [deg]
frequency
3
Accumulated phase
2 angle difference
contained within
1 7deg.
0
Accumulated angle
resets if df<0.002Hz
-1
500 550 600 650 for more than 2s.
t [s]
Satellite based LOM protection
reasonable margin
6 (16deg), i.e. GPS
accuracy in time
4 synchronisation is
2 not essential.
0
-2
-4
500 550 600 650
t [s]
Internet based LOM protection
General WAM system for LOM
Laverty, D.M.; Morrow, D.J.; Best, R.J.; Crossley, P.A.; , "Differential ROCOF relay for Loss-of-Mains protection of Renewable Generation
using phasor measurement over Internet Protocol," Integration of Wide-Scale Renewable Resources Into the Power Delivery System, 2009
Reverse VAR Method
Reverse VAR relay measures the generator reactive
power flow Qgen and operates when it exceeds a fixed
threshold.
Trip if Qgen > Qs
LOM LOM
G G
Pgen Pgen
Qgen0
Connected Qgen Islanded
Centralised LOM protection using IEC-61850
The technique is based on communication of CB status
to the central controller.
D
CB1 CB3
C CB4
CB2 CB2
CB1 CB2
CB1 CB5
CB3
E
MAINS
CB3
CENTRAL
CONTROLLER
F G
CB4 CB5
CB3 CB4
Coffele, F.; Moore, P.; Booth, C.; Dysko, A.; Burt, G.; Spearing, T.; Dolan, P., Centralised Loss of Mains protection using IEC-61850,
Developments in Power System Protection (DPSP 2010).
Reverse VAR Method
Interconnected Psys=0
system
LOM Qc
G Qc
Pgen Qc
Qgen
Intertripping of Open Ring Feeders
Signal send
Signal block
G G
Signal receive Open point
Signal receive
Signal block
Signal send
G
Signal receive
Intertripping of Open Ring Feeders
Signal send
Signal block
G G
Signal receive Open point Signal receive
Signal block
Signal send
G
Signal receive
Active methods