Transcript System Separation

“Cascading Events
and
How to Prevent Them”
The International Meeting of VLPGO
WG#1
October 25, 2005
1
Background
Recurrent Cascading Outage Worldwide after Market Liberalization
1st International Meeting of Very Large Power Grid Operators(VLPGO)
- October 25-26, 2004, - Philadelphia, USA
- Objectives :
To find and share common concerns in “Maintaining Reliability”
To exchange good/bad experiences and the best existing practices
To cooperate on developing necessary measures
Set up three working groups
WG#1 :Cascading Events and How to Prevent Them (Lead: TEPCO)
WG#2 :EMS Architectures for 21th Century
(Lead : PJM/MISO)
WG#3 :Advanced Decision Support Tools
(Lead : RTE)
2
Objectives and Means
WG-1 surveyed the ways to prevent
cascading events through a questionnaire
to identify:
Common causes and mechanisms
Existing measures to be recommended
New technologies to be developed in the
future
3
Contributors to WG#1
Mr.Hideaki TANAKA
(Coordinator)
Mr.Masanobu KAMINAGA
TEPCO
JAPAN
TEPCO
JAPAN
Mr.Michel KORMOS
PJM
USA
Dr.Yuri MAKAROV
CAISO
USA
Mr.Temistocle BAFFA SCIROCCO GRTN
ITALY
Mr.Jean Michel TESSERON
RTE
FRANCE
Mr.Ian WELCH
National Grid UK
4
Contents
1. Scope of Survey
2. Survey Results
3. Conclusion
5
1. Scope of Survey
6
Power System States and Transition
Security Monitoring
Normal
Restoration
Preventive
Control
Restorative
Alert
Emergency
Control
In extremis
Emergency
7
Scope of each WG
WG-1
De-Centralized
System
Monitoring
Centralized System
< In Control Centers >
WG-3
*New technologies
used by operators
for decision
making
WG-2
*Standardization of
architecture
platform for EMS
Control
*Three WGs focused on technical issues excluding
institutional issues, such as power market design.
8
Issues associated with Cascading Events
Cascading Blackouts
Power System
Thermal
Stability
Overloading
Rotor Angle
Frequency
Voltage
Stability
Stability
Stability
Small-Disturbance
Transient
Large-
Small-
Angle Stability
Stability
Disturbance
Disturbance
Voltage Stability
Voltage Stability
9
2. Survey Results
10
Features of Very Large Power Grid
Organization
TEPCO
System
Capacity
[GW]
64
Highest
Voltage
[kV] NG
★
500
★
GRTN
54
380
National Grid
54
400
PJM
133
765
CAISO
48
500
RTE
83
400
RTE
CAISO
★
GRTN
★
★
PJM
★
TEPCO
11
System Configuration and Interconnection with
Neighboring Systems
System Configuration
Organization
EHV
HV,MV
Interconnections
TEPCO
Mesh
[ 63 kA]
Radial
A 500kV AC link
Two BTBs
GRTN
Mesh
[ 50 kA]
Radial
Eight 380kV AC links
A 500kV DC link
A 200kV DC link
National Grid Mesh
[ 63 kA]
Radial
AC Link
DC link
PJM
Mesh
[ 63 kA]
Mesh
248 AC links
CAISO
Mesh
[ 63 kA]
Mesh
Five AC links
RTE
Mesh
[ 63 kA]
Mesh
40 AC links
A 270kV DC link
[Maximum Short Circuit Current]
12
Power Flow Level and Critical Stability Issues
Maximum
Power
Flow/ SIL
Critical Issues
Smalldisturbance
2.0
1.5
○
(3)
○
CAISO
RTE
(Priority)
○
(2)
Voltage
○
(2)
○
(1)
○
National
Grid
PJM
Frequency
○
(1)
TEPCO
GRTN
Transient
Thermal
Overloading
○
(3)
○
(2)
○
○
○
○
○
○
○
(3)
○
(1)
○
13
Overview of Past Cascading Events
<20th Century: Before Market Liberalization>
Date/
Area
MW Lost,
Duration
Causes
(Trigger)
Nov. 1978/ 30,000MW - Major 400kV lines trips
due to overloading
France
7 hr
Aug. 1981/ 1,900MW
UK
2.5 hr
- Loss of 3 400kV circuits
and lower voltage
interconnection
- Multiple generator trips
Jan. 1987/
France
8,000MW
3 hr
Jul. 1987/
Tokyo
8,000MW
4 hr
Jul./Aug.
1996/
Western
US
11,850MW - Loss of multiple lines
and loss of critical
? hr
generation
28,000MW
9 hr
- Insufficient VAR supply
for high rate of load
pickup
Critical
Phenomenon
-Voltage collapse
-System
separation
-System
separation
-Voltage collapse
-Voltage collapse
-Voltage collapse
-System
separation
-Power oscillation
14
Overview of Past Cascading Events
<21th Century: After Market Liberalization>
Date/
MW Lost,
Causes
Duration
Area
(Trigger)
Aug. 2003 61,800MW -Multiple line trips
Northern
42 hr
US
Critical
phenomenon
-Voltage collapse
-System separation
Aug. 2003
London
-Overloading
724MW -Incorrect operation of a
backup relay
0.7 hr
Sep. 2003 20,000MW -Multiple EHV line trips
Italy
20 hr
Sep. 2003
Scandia
6,550MW -Scrum of a Nuke plant
6.5 hr -Double-bus fault
May. 2005
Moscow
2,500MW
32 hr
-CT explosion
-Multiple Tr explosion&fire
-Voltage collapse
(before system
separation)
- Frequency collapse
(after system
separation)
-Power oscillation
-System separation
-Voltage collapse
-Power oscillation
-Overloading
15
Common Causes of Recent Cascading Outages
<Cause> Deregulation of Electricity Market: Monopoly to Competition
Less Maintenance
Priority to Market
Mechanism
Change of System Operation Rules
More Players
More..
Complicated and
Enlarged Power Grid
Information to be
Handled
Delay in Network
Enhancement
Increase of Inter-regional (National)
Power Exchange
Uncertainty in Operating Condition
Fault
Frequency
<Impact>
(To Interconnected Power System)
Heavier Duty in Accommodating
Electricity Transaction
(To Operators)
Difficulty in Responding to
Abnormal Situation
16
Mechanism of Cascading Outages-1
Trigger
-Unexpected heavy loading
Impact of Market Liberalization
-Unscheduled generation outage
Cost Reduction
-Single fault (tree touching etc.)
Wide-area Heavy Power
Transaction
-Combination of above events
Delay in Initial Action
-Delay in grasping the situation
-Delay in communication with
neighboring operators
-Delay in taking mitigation action
Inappropriate On-line Monitoring
Need for Sophisticated On-line
Monitoring System
Need for On-line Contingency
Analysis Tool
Need for Automatic Preventive
Control
17
Mechanism of Cascading Outages-2
Cascading Events: Alert to Emergency
-Overloading
-Voltage Collapse
-Power Oscillation
-Loss of Synchronism
-Frequency Declining
Need for Enhancement of
Emergency Control, so called
‘Safety Net,’ including ‘Islanding’
Protective Action
-Transmission Lines
-Transformers
-Generators
Need for Time Coordination with
Safety Nets, such as UFLS and
UVLS
18
Existing Countermeasures (Security Monitoring)
<Findings>
1) On-line state estimator(SE) is commonly used
*Maximum Capacity: 7400 nodes, 2500 generators, every one minute
2)On-line contingency analysis is also commonly implemented, both for
‘voltage instability’ and ‘thermal overloading’
3) On-line direct monitoring of power system oscillation has started in some
countries. In the US and EU, GPS-based PMU (Phasor Measurement
Unit) is applied to monitor the phase.
<Challenges>
1) To improve the accuracy of SE by using the PMU in combination with the
conventional SE
2) To extend the scope of contingency evaluation into phase angle stability,
in particular transient stability
3) To estimate the frequency/power regulation performance of the system
19
Existing Countermeasures (Preventive Control)
< Findings>
1) Power System Stabilizer is a common tool to prevent small-signal
instability.
2) High-speed re-closing including multi-pole re-closing is adopted in order
to improve transient stability, while aiming to put the network back to its
initial state.
3) Automatic Generator Control is commonly used to maintain system
frequency.
4) A wide variety of de-centralized automatic control systems are used to
prevent the transition to the alert state, as well as to lighten operators’
burden.
5) In France, a centralized automatic control system called the Secondary
Voltage Control System, has been in operation at the regional level, the
purpose of which is to get better control performance.
<Challenges>
1) To select a centralized system or de-centralized system appropriately in
accordance with system features
2) To develop a sophisticated algorithm that can provide the operators with
information on how to prevent the transition from “Alert” to “Emergency”.20
Existing Countermeasures(Emergency Control)
< Findings>
1) Several types of ‘emergency controls’ have been developed as a
‘safety net’ and are in operation.
They are categorized into the following three categories:
<Generator Tripping>
OFLS, SPS (Generators, Pumped-Storage Units etc.)
<Load Shedding>
UFLS, UVLS, SPS (Loads), [Blocking of Tap Changers]
<System Separation>
Islanding
<Challenges>
1) To keep the interconnection as long as possible, even when an
emergency occurs, ‘Time coordination’ between the equipment
protection system and emergency control systems must be examined.
2) To consider quicker restoration when designing the safety net.
21
Future Countermeasures (Under Development)
<Findings>
1) The following systems, mainly for “Monitoring “, are now
under development.




On-line transient stability assessment in the US and Japan
PMU-based small-signal stability monitoring system in the US.
A wide area monitoring system, which covers all issues regarding
cascading events and includes the centralized voltage control
system in Italy.
Monitoring of Generator performance and of f/P power system
performance in France.
2)Currently, no emergency control system is being developed.
22
Other Issues
1) Institutional Measures
◆Give TSO operators more authority
◆Establish a reliability standard applied to all relevant
stakeholders
◆Contract among stakeholders who comply with the reliability
standard
◆Establish strong coordination between TSOs in different
time frames
2)Other Measures
◆ Aim Operator Training Simulator at:
Improving the knowledge and skills of the individuals
Developing the operator’s tolerance for psychological
stress and capability of coping with abnormal conditions
◆Risk indices used to set adequate reserve margins or to
allow operators know the necessity of load shedding
23
3. Conclusion
24
Summary (Existing Measures)
1)There were several types of automatic
centralized/de-centralized systems for “Monitoring”,
“Preventive Control”and “Emergency Control”,
respectively.
<centralized> <decentralized>
Monitoring
:
◎
Preventive Control :
○
○
Emergency Control :
◎
2) Each VLPGO will be able to select and employ the
most appropriate systems from the “Seasoned Best
Practice Menu”.
25
Summary (Future R&D Topics)
1)Since the use of “Emergency control ” is a
last resort measure, we should focus future
R&D mainly on “Monitoring ” and
“Preventive Control ”, which are tools used
“upstream” of cascading events.
2)Upon reviewing survey results, we have
identified the following R&D topics to be
further addressed.
26
< Reinforcement of Monitoring>




On-line Dynamic Assessment
(PJM,TEPCO)
- On-line High Speed Screening
On-line System stability(Steady state/Dynamic)
Monitoring
(CAISO)
- PMU application
- Eigenvalue calculation
Wide Area Measurement
(GRTN)
Generator and System f/P Performance
Monitoring(RTE)
<Reinforcement of Preventive Control>

On-line Corrective Action(switching or re-dispatching)
Indicating System
(RTE)
27
3) In the process of R&D, we should take
into account :




Up-to-date IT technologies
Parallel computing techniques
Advanced algorithms for on-line analyses
Standardization of software and architecture.
28
Recommendation on 2006 action plan
<Mission>
Develop a “Comprehensive Survey Paper” on
the ways to prevent cascading blackouts by
the next VLPGO meeting for presentation at
an International Conference such as CIGRE
<Action>
1. Recruit new members and dispatch
questionnaire.
2. As a new and last aspect, add a “Restoration”.
3. Extend the survey area to the papers
published by the PGOs with a capacity of less
than 50GW.
29
Thank you for your attention
30