what do fault statistics tell us regarding causes ......risk and vulnerabilities in power systems in...

Workshop DNV – NTNU 2011-09-27 Risk and vulnerabilities in power systems in light of climate change 1

What do fault statistics tell us regarding causes resulting in power outages?

Gerd H. Kjølle

Professor II Dept. of Electric Power Eng., NTNU Senior Research Scientist, SINTEF Energy Research


Content of presentation

FASIT – The Norwegian standard for reliability data collection and reporting

Faults and interruptions (power outages) – definitions Highlights from the fault statistics 1 – 420 kV

Number of events and energy not supplied 1989 – 2010 Fault causes 2007 – 2010 Component faults and fault rates

Interruptions and cost of energy not supplied (CENS) Large disturbances (high impact) Brief comparison Nordic countries

Workshop DNV – NTNU 2011-09-27 Risk and vulnerabilities in power systems in light of climate change

Standard for collection, calculation and reporting – FASIT

Introduced in 1995 Used by all network

companies in Norway 6 software vendors Software quality

assurance (contracts and acceptance test)

FASIT software

Guidelines for data

collection

Reporting schemes

≥ 33 kV 1 - 22 kV

< 1 kV

Software specification

Basic requirements

www.fasit.no


FASIT – reliability data classes

Fault

Failure cause

• Component

• Voltage level

Interruption

No interruption

Forced outage

Maintenance

Planned disconnection

CENS

• ENS • Affected end-user groups • Responsible company

• End-user

• Network level/type


Interruption – definition (EN 50160)

Condition in which the voltage at the supply terminals is lower than 5 % of the reference voltage.

A supply interruption can be classified as prearranged, when network users are informed in advance,

or accidental, caused by permanent or transient faults, mostly

related to external events, equipment failures or interference. An accidental interruption is classified as: a long interruption (longer than 3 min); a short interruption (up to and including 3 min)

Power outage = interruption in this presentation


Fault – definition

Fault is the state of an item characterized by inability to perform a required function (IEC)

Fault causes may be related to construction, production, installation, use or maintenance causing fault on the unit

Fault causes may be classified in triggering, underlying or contributing causes

Faults are divided in Permanent (corrective maintenance/repair) Transient/ temporary (no corrective maintenance/repair,

reconnection of breaker or replacement of fuse)


Power system levels – Norway

420 kV 300 kV

132 kV 66 kV

22 kV 11 kV

230 V 400 V

Import/Export Import/Export

Generation Transmission Sub-transmission Distribution

G

G

G

G


Number of events 1 – 420 kV

About 25000 events per year > 95 % in the distribution network, ~ 50 % disturbances

0

5000

10000

15000

20000

25000

30000

35000

Planned disconnections 33 - 420 kV

Disturbances 33 - 420 kV

Planned disconnections 1 - 22 kV

Temporary failures 1 - 22 kV

Permanent failures 1 - 22 kV

FASIT introduced Lack of detailed data for 2006


Energy not supplied 1 – 420 kV

~ 80 % caused by faults in the distribution network

0

10000

20000

30000

40000

50000

60000

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

ENS due to planned disconnections 33 - 420 kV (from 2001 only)

Sum

ENS due to disturbances 33 - 420 kV

ENS due to planned disconnections 1 - 22 kV

ENS due to disturbances 1 - 22 kV

Lack of detailed data for 2006 Sum 2007 should be 14800 MWh

Revenue regulation CENS

MWh per year


Triggering fault causes 1 – 22 kV 2007 - 2010

10

Operational stress3 %

No triggering fault cause2 %

Construction/maintentance etc

2 %Human errors

6 %

Environment43 %

Technical equipment11 %

Previous faults0 %

Cause not clarified28 %

Unknown5 %

Disturbances 1-22 kV divided in triggering fault cause 2007-2010

No of disturbances: Energy not supplied (ENS):

Operational stress4 % No triggering fault

cause2 % Construction/

maintentance etc3 %

Human errors7 %

Environment45 %

Technical equipment22 %

Previous faults0 %


Unknown3 %

ENS due to disturbances 1-22 kV divided in triggering fault cause 2007-2010

34 %

12 %

7 %1 %

18 %

20 %

8 %

Lightning

Wind

Snow/ice

Salt/pollution

Vegetation

Birds/animals

Other/unknown

18 %

15 %

14 %2 %

36 %

5 %

10 %

Lightning

Wind

Snow/ice

Salt/pollution

Vegetation

Birds/animals

Other/unknown

Weather and unknown/not clarified fault causes dominate


Triggering fault causes 33 – 420 kV 2007 – 2010

11

No of disturbances: Energy not supplied (ENS):

Operational stress6 %


7 %

Human errors15 %

Environment42 %

Technical equipment

18 %

Previous faults0 %


Unknown2 %

Distburbances 33-420 kV divided in triggering fault cause 2007-2010Operational stress

4 %


24 %

Human errors15 %

Environment27 %

Technical equipment

22 %

Previous faults1 %


Unknown1 %

ENS due to disturbances 33-420 kV divided in triggering fault cause 2007-2010

50 %

16 %

8 %

3 %

12 %

3 %8 %

Lightning

Wind

Snow/ice

Salt/pollution

Vegetation

Birds/animals

Other/unknown

34 %

22 %9 %

3 %

22 %

2 %8 %

Lightning

Wind

Snow/ice

Salt/pollution

Vegetation

Birds/animals

Other/unknown

Weather , technical equipment and human errors dominate


Component faults 1 – 22 kV 2007 – 2010

12

0

500

1000

1500

2000

2500

3000

3500

0

500

1000

1500

2000

2500

3000

3500

4000

4500

Overhead line Underground cable Breaker, disconnector Transformer Other Not identified

No

of fa

ults

per

year

MW

h pe

r yea

r

1-22 kV

No of faults per year 2007-2010

ENS per year år 2007-2010


Component faults 33 – 420 kV 2007 – 2010

13

0

100

200

300

400

500

600

700

800

900

1000

0

50

100

150

200

250

300

Overhead line Underground cable Breaker, disconnector

Transformer Other Not identified Protection Control equipment

MW

h pe

r yea

r

No

of fa

ults

per

year

33-420 kV

No of faults per year 2007-2010

ENS per year 2007-2010


Fault rate for overhead lines 1 – 22 kV 1989 – 2005

0,0

2,0

4,0

6,0

8,0

10,0

12,0

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Year

No.

of f

aults

per

100

km

Temporary faultsPermanent faultsAll faults

FASIT introduced

New version of FASIT


Fault rate overhead lines 1 – 22 kV 2007 – 2010

15

0

1

2

3

4

5

6

7

8

2007 2008 2009 2010

Faul

ts p

er 1

00 km

per

yea

r

Fault frequency overhead line 1-22 kV

Temporary Permanent All faults

1 – 22kV: 6 – 7 faults per 100 km per year



16

0,0

0,5

1,0

1,5

2,0

2,5

3,0

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

No o

f fau

lts p

er 1

00 k

m p

er y

ear

År

Fault frequency 33 - 420 kV

420 kV

220 - 300 kV

132 kV

33 - 110 kV



17

420kV – 33 kV: ~ 0,5 - 1 faults per 100 km per year

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8

2007 2008 2009 2010

No

of fa

ults

per

100

km p

er ye

ar

År

Fault frequency overhead lines 33-420 kV420

300-220

132

110-33


Interruptions 2005 – 2010, long interruptions > 3 minutes

18

0

0,5

1

1,5

2

2,5

3

3,5

2005 2006 2007 2008 2009 2010

No

of in

terr

uptio

ns p

er y

ear

Long interruptions

No of interruptions per delivery point

SAIFI

0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

2005 2006 2007 2008 2009 2010

Hour

s per

year

Long interruptions

Annual interruption duration per delivery point

SAIDI

About 3 long interruptions and 3 – 4 hours per year per delivery point


Interruptions 2006 – 2010, short interruptions ≤ 3 minutes

19

About 2,5 short interruptions and 2 minutes per year per delivery point

0

0,5

1

1,5

2

2,5

3

3,5

2005 2006 2007 2008 2009 2010

No

of in

terr

uptio

ns p

er y

ear

Short interruptions

No of interruptions per delivery point

SAIFI

0

0,5

1

1,5

2

2,5

3

2005 2006 2007 2008 2009 2010

Min

utes

per

yea

r

Short interruptions

Annual interruption duration per delivery point

SAIDI


Cost of energy not supplied caused by different system levels

20

0

100000

200000

300000

400000

500000

600000

2007 2008 2009

NO

K 10

00 Distribution

Regional grids

Main grid

Total

Cost of short interruptions included from 2009

Distribution counts for 78 %


Normal/frequent events vs major events (large disturbances, HILP)

Power system failures

Causes Consequences Natural hazard Technical/ operational Human errors Terror, sabotage etc.

Barriers

Minor Moderate

Major Critical Catastrophic

Wide-area interruption/

Blackout

Fault statistics mainly give information about normal/frequent events


Examples of large disturbances (blackouts), Nordic, Europe, Canada

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0,1 1 10 100 1000Hours

Dis

conn

ecte

d lo

ad (M

W)

Sweden 1983

Southern Sweden/ EasternDenmark 2003Western Norway 2004

Helsinki 2003

Southern Sweden 2005(Gudrun)Canada 1998

France 1999

Europe, UCTE 2006


More examples, Norway, US/Canada...

Historic blackouts: Duration and disconnected load

1

10

100

1.000

10.000

100.000

0,1 1 10 100 1000

Stipulated average duration (log h)

Dis

conn

ecte

d lo

ad (l

og M

W)

Gudrun '05

Steigen, Norway '07

Oslo S '07

US /Canada '03

W. Norway '04

Swe/DK '03

Canada '98

Project Vulnerability and security in a changing power system, Nfr/SINTEF Energi, 2009 - 2012

Critical infrastructures

Energiutvalget 25. august 2011


Fault causes major events - examples

”Western Norway”, February 2004, 300 kV

Breakage of line joint Delayed protection response

Causes: Construction fault Degradation of components ”Inappropriate” protection

Steigen, January 2007, 66 kV Breakdown of both overhead

lines Causes :

Storm, icing Construction fault Degradation (ageing)


Comparison Nordic countries

25

ENTSO-E Nordic Grid disturbance and fault statistics 2008



26




27



Extra slides


Software certification: FASIT test network

LP2

D ~ G1

10/66 kV

BE1

RP1

BE2 BE3

SA1

KL1 KL2

BE4 BE5

BE7

66/22 kV T2

A

C

BE8

KL3

BE6

SA2

LP1

B

BS1

SA4

BE10 T3

RP2

BE9

KL4

BS2

SA5 BE11 T4

RP3

KL5 SA6

BL7 T5

RP4

KA2

SA7

BE13 T6

RP5

KA3 SA8 BE14 T7

RP6

SA3

KA1

SONE 2

BE17

BE15

BS5

BS6

KL6

BS3 KA4 BS4

T1

BE16

E

SONE 1

BS30

G: Generator

SA: Busbar

BE: Circuit breaker

BS: Disconnector

KL: Overhead line

KA: Underground cable

T: Transformer

RP: Delivery (load) point

LP: Delivery point

A, B, C, D: Network companies

Workshop DNV – NTNU 2011-09-27 Risk and vulnerabilities in power systems in light of climate change 30 30

U.S. and Canada, August 14, 2003

Ref.: Final Report on the 2003 Blackout in the United States and Canada: Causes and Recommendations, U.S.-Canada Power System Outage Task Force

Simplified description of the event

1. 12.15-14.54: Malfunctioning software systems limiting the operators situation awareness and control (State Estimator, SCADA alarm and logging, EMS terminals and server)

2. 13.31: Trip of important generation increases loading on lines

3. 14.02-16.05: Tripping of highly loaded lines, with premature tripping of many lines due to inadequate vegetation management

4. 16.06-16.11: This eventually caused instability, triggering cascaded tripping, separating the Eastern Interconnection into two asynchronous areas

5. 16.11-16.13: Large differences between load and generation, led to instability and blackout of the island consisting of parts of Northeastern U.S and Ontario

Time between triggering event and cascaded tripping ~ 4 h

Extraordinary Events – Understanding Sequence, Causes, and Remedies E. Johansson




Threats Malfunction of computer

systems for system operation Overgrown vegetation Inadequate system

understanding, operator training and clarification of responsibility

Inadequate protection system/ scheme

Final consequences for end-users 50 million people affected 61 800 MW lost, 350 000 MWh

lost

US / Canada, August 14 2003 Overview of the course of events.1

The map highlights the affected regions.

Aug 14 System in normal state, within prescribed limits. High, but not abnormally high loads

12:15 Failure (info) Erroneous input data put the Midwest Independent System Operator’s state estimator and real time contingency analysis tool out of service

13:31 Failure Generator at Eastlake power plant trips – loss of important source of reactive power

14:02 Failure 345 kV line trips due to tree contact caused by high temperature and line-sagging

14:14 Failure (info) Control room operators at First Energy looses the alarm function (with no one in the control room realising this)

15:05 – 15:41 Failures

Three 345 kV lines into the Cleveland-Akron area trips due to tree contact. Loads on other lines increase

15:42 Operators at First Energy begin to realise that their computer system is out of order and that the network is in serious jeopardy.

Failures Decreased voltage and increased loading of the underlying 138 kV system, causes 16 lines to fail in rapid order

16:06 Failure Loss of the 345 kV Sammis-Star line between eastern and northern Ohio due to overload. Triggers the cascade

Cascade

Uncontrolled power surges and overload causes relays to trip lines and generators. Northeastern US and Ontario form a large electrical island, which quickly becomes unstable due to lack of generation capacity to meet the demand. Further tripping of lines and generators breaks the area into several electric islands, and most of these black out completely. Some smaller islands with sufficient generation manage to stabilize.

16:13 Cascade over. 50 million people deprived of power

Aug 15 Restoration Approx 80 % of the energy restored

Aug 22 Restoration completed




Vulnerabilities Lack of sufficient tools, competence and standards, leading to: Inadequate situational awareness Insufficient diagnostic support from the interconnected grid’s reliability

coordinator (MISO) Barriers to prevent component failure

Vegetation management Monitoring of lines and operation to prevent overload.

Barriers to prevent power system failure Situation awareness and response of TSOs, operator training Computer tools for monitoring, and back-up systems for these Reliability standards and clear areas of responsibility; ensure operation

within secure limits Information sharing between TSOs

What do fault statistics tell us regarding causes resulting in power outages?Content of presentationStandard for collection, calculation and reporting – FASIT FASIT – reliability data classesInterruption – definition �(EN 50160) Fault – definition Power system levels – Norway Number of events 1 – 420 kV Energy not supplied 1 – 420 kVTriggering fault causes 1 – 22 kV �2007 - 2010Triggering fault causes 33 – 420 kV 2007 – 2010 Component faults 1 – 22 kV �2007 – 2010Component faults 33 – 420 kV �2007 – 2010Fault rate for overhead lines 1 – 22 kV 1989 – 2005 Fault rate overhead lines 1 – 22 kV 2007 – 2010 Fault rate overhead lines 33 – 420 kV 1989 – 2005 Fault rate overhead lines 33 – 420 kV 2007 – 2010 Interruptions 2005 – 2010, �long interruptions > 3 minutes Interruptions 2006 – 2010, �short interruptions ≤ 3 minutes Cost of energy not supplied caused by different system levelsNormal/frequent events vs major events (large disturbances, HILP)Examples of large disturbances (blackouts), Nordic, Europe, CanadaMore examples, Norway, US/Canada...Fault causes major events - examplesComparison Nordic countriesComparison Nordic countriesComparison Nordic countriesExtra slidesSoftware certification: FASIT test networkU.S. and Canada, August 14, 2003U.S. and Canada, August 14, 2003U.S. and Canada, August 14, 2003

what do fault statistics tell us regarding causes ......risk and vulnerabilities in power systems in...

Documents