Reliability Performance of Electric Utilities

presenterbelchManagement

Nov 18, 2013 (3 years and 6 months ago)

59 views

1

A Comparable Method for Benchmarking the
Reliability Performance of Electric Utilities


Dan Kowalewski

630
-
437
-
2905

daniel.kowalewski@exeloncorp.com



IEEE PES 2002 Summer Meeting

Chicago, Illinois

Palmer House Hilton


Private Dining Room 9

Monday, July 22, 2002


2:00 to 5:00 PM


2

Overview


Criteria and Desired Elements


Analysis


Approach


Alternatives Considered


Proposed Application Rules


Calculation Specifics


Analysis Results


Conclusions


3

Event Criteria


Create Event Exclusion Criteria


Facilitates Consistent Exclusions


Allows “Fair” Benchmarking


Provides Information to Understand Performance Trends


Repeatability of Results

4

Desired Elements


Has Clear Rules for Application and Exclusion


Removes Subjectivity


Scales to Various Sized Utilities


Removes Outlying Events


Removes Major Weather Variability


Allows Trends to Appear


Balances Operational Concerns


Frequency


Duration


Allows for Consistent Results


Ease of Calculation

5

Analysis: Approach


Compare Proposed Methods to Understand


Consistency and Understandability


Differences of Exclusions


Impacts on Different Sized Companies


Variability of Results


A Great Deal of Confusion Initially


Methodology and Calculation Approach was Not Clear



3 Years vs 5 Years of Historic Data


Log vs Non
-
Log Data


Basis Unclear: SAIDI, SAIFI, Customer Minutes


Had to Apply Data to Understand


Traded Methodology and Results


Ensures Consistent Application


Expanded Understanding with Broader Data Set


6

Analysis: Alternatives Considered


Traditional 10% IEEE Definition


ComEd Region


ComEd


PECO


Bootstrap


3 Year Historic


5 Year Historic


2.5 or 3 Beta (SAIDI per Day)


3 Year Historic


5 Year Historic


6 Beta (SAIDI per Day)


Other Analysis


SAIDI per Calendar Day



Start Time of Interruption


Customer Interruption Minutes Experienced per Day


Rolling 24 Hour Window

7

Alternative Comparison

10%
Sys/Region
x
x
x
Boot Strap
x
x
x
x
x
2.5 or 3 Beta
x
x
x
x !
x
x
x
x
x
6 Beta
x

x
x
8

ComEd Major Event Days

METHOD
1998
1999
2000
2001
Exelon Energy
11
8
10
16
10% Regional
3
0
2
0
10% System
2
0
0
0
Bootstrap
7
2
3
0
3 Beta (SAIDI / Day)
10
3
5
0
3 Beta (SAIDI / Day)
10
2
6
0
3 Beta (SAIDI / Day)
17
4
11
0
6 Beta (SAIDI / Day)
0
0
0
0
Hypothetical Application of 1998 - 2001 Threshold Values
*

* * *

*

Based On Outage Start Date

* *

Customer Interruption Minutes by Calendar Day

* *

* * *

Based on 24 hour rolling customer outages

9

ComEd Major Event Days

METHOD
Exelon Energy
10% Regional
10% System
Beta (SAIDI / Day)
2.5
3
2.5
3
2.5
3
2.5
3
Major Event Days
15
10
6
3
7
5
6
0
6 Beta (SAIDI / Day)
0
0
0
0
2000
2001
11
8
10
16
1998
1999
Hypothetical Application of 1998 - 2001 Threshold Values
3
0
2
0
2
0
0
0
*

*

Based On Outage Start Date

10

3 Beta SAIDI Day
-
Threshold Calculation



IEEE data (all outages > 5minutes)


Five years of most recent historic SAIDI per day data


Take the log of SAIDI per day values


Calculate the mean


Calculate the population standard deviation


Calculate the mean plus 3 times the standard deviation


Convert back to SAIDI per day values by taking the exponential
of the calculated mean plus 3 times the standard deviation
number


Results in a Threshold Exclusion value


Threshold value is applied to the year following the five
-
year data
analysis period


All days with SAIDI greater than or equal to the calculated
Threshold value are considered Major Event Days



11

8.13
9.25
7.27
8.30
8.51
7.07
7.50
8.46
5.25
8.88
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.00
1995
1996
1997
1998
1999
2000
2001
2002
YEAR
SAIDI(MIN) PER DAY
PECO
COMED
Thresholds of Daily SAIDI

(3 Year Rolling)

12

Thresholds of Daily SAIDI

(5 Year Rolling)

7.53
8.45
7.77
7.06
7.70
6.07
8.08
7.78
7.97
8.58
5.93
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.00
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
YEAR
SAIDI(MIN) PER DAY
PECO
COMED
(4 yr average)
13

Variability of Thresholds

ComEd & PECO
2.5 and 3 Beta versus Bootstrap Daily SAIDI Thresholds
0
5
10
15
20
25
30
35
40
45
50
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
SAIDI(MIN) PER DAY
ComEd Bootstrap 1 - 4 yrs
ComEd 3 Beta 1 - 4 yrs
PECO 3 Beta 5 yrs
PECO Bootstrap 5 yrs
ComEd 2.5 Beta 1 - 4yrs
PECO 2.5 Beta 5 yrs
14

SAIDI/Day Thresholds
5 yrs of Data
0
2
4
6
8
10
12
14
16
18
20
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
SAIDI(MIN) PER DAY
SAIDI/Day Thresholds
3 yrs of Data
0
2
4
6
8
10
12
14
16
18
20
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
SAIDI(MIN) PER DAY
Bootstrap Thresholds can be highly volatile, especially with
few years of data (the saturation effect).

PECO
-

Variability of Thresholds

3 Beta

Bootstrap

Bootstrap

3 Beta

15

ComEd Daily SAIDI (1998
-

2001)

ComEd Daily SAIDI (1998-2001)
07/31/99
07/30/99
07/19/98
06/18/98
04/08/00
03/10/98
05/28/98
06/26/98
09/07/98
08/24/98
11/10/98
03/09/98
09/11/00
05/18/00
08/06/00
05/08/00
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
110.0
120.0
130.0
140.0
150.0
160.0
170.0
180.0
190.0
200.0
1/1/98
2/1/98
3/1/98
4/1/98
5/1/98
6/1/98
7/1/98
8/1/98
9/1/98
10/1/98
11/1/98
12/1/98
1/1/99
2/1/99
3/1/99
4/1/99
5/1/99
6/1/99
7/1/99
8/1/99
9/1/99
10/1/99
11/1/99
12/1/99
1/1/00
2/1/00
3/1/00
4/1/00
5/1/00
6/1/00
7/1/00
8/1/00
9/1/00
10/1/00
11/1/00
12/1/00
1/1/01
2/1/01
3/1/01
4/1/01
5/1/01
6/1/01
7/1/01
8/1/01
9/1/01
10/1/01
11/1/01
12/1/01
Date
SAIDI
Non-Storm Days
Storm Days
2.5 BETA = 4.3 (34 days)
3 BETA = 7.70 (18 days)
10% IEEE (System)
2 days
10% IEEE (Region)
5 days
16

ComEd Daily SAIDI (1998
-

2001)

ComEd Daily SAIDI (1998-2001)
07/31/99
07/30/99
07/21/99
08/25/98
07/19/98
06/18/98
09/11/00
04/08/00
05/08/00
05/28/98
06/26/98
09/07/98
03/10/98
08/06/00
05/18/00
08/24/98
11/10/98
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
1/1/98
2/1/98
3/1/98
4/1/98
5/1/98
6/1/98
7/1/98
8/1/98
9/1/98
10/1/98
11/1/98
12/1/98
1/1/99
2/1/99
3/1/99
4/1/99
5/1/99
6/1/99
7/1/99
8/1/99
9/1/99
10/1/99
11/1/99
12/1/99
1/1/00
2/1/00
3/1/00
4/1/00
5/1/00
6/1/00
7/1/00
8/1/00
9/1/00
10/1/00
11/1/00
12/1/00
1/1/01
2/1/01
3/1/01
4/1/01
5/1/01
6/1/01
7/1/01
8/1/01
9/1/01
10/1/01
11/1/01
12/1/01
Date
SAIDI
Non-Storm Days
Storm Days
(RESCALED)
2.5 BETA = 4.30 (34 days)
3 BETA = 7.70 (18 days)
10% IEEE (System)
2 days
10% IEEE (Region)
5 days
17

ComEd Daily SAIDI (1998
-

2001)

ComEd Daily SAIDI (1998-2001)
06/06/99
11/11/98
10/25/01
08/09/01
07/23/01
07/22/01
06/12/01
04/07/01
10/08/00
05/09/00
05/17/99
06/11/99
06/28/98
07/20/98
06/25/98
05/29/98
04/08/00
07/21/99
08/25/98
07/19/98
06/18/98
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
1/1/98
2/1/98
3/1/98
4/1/98
5/1/98
6/1/98
7/1/98
8/1/98
9/1/98
10/1/98
11/1/98
12/1/98
1/1/99
2/1/99
3/1/99
4/1/99
5/1/99
6/1/99
7/1/99
8/1/99
9/1/99
10/1/99
11/1/99
12/1/99
1/1/00
2/1/00
3/1/00
4/1/00
5/1/00
6/1/00
7/1/00
8/1/00
9/1/00
10/1/00
11/1/00
12/1/00
1/1/01
2/1/01
3/1/01
4/1/01
5/1/01
6/1/01
7/1/01
8/1/01
9/1/01
10/1/01
11/1/01
12/1/01
Date
SAIDI
Non-Storm Days
Storm Days
(RESCALED)
2.5 BETA = 4.30 (34 days)
10% IEEE (System)
2 days
10% IEEE (Region)
5 days
3 BETA = 7.70 (18 days)
18

Conclusions


Recommend 3 Beta SAIDI per Day


Achieves Objectives


Consistent


Facilitates Fair Benchmarking


Provides Information to Understand Performance
Trends


Repeatability of Results


Proven with
REAL

Data