SRP Voltage Sag Index Methodology

Experience and FAQs

Rao Thallam and Kris Koellner

IEEE PES P1564 TF Meeting
Las Vegas, 1/28/2003

IEEE PE Review July 2001

!

While most utilities can quote

reliability numbers ..; they
cannot provide a measure of
performance for the quality of
power supply
Don Volzka, PES President
2

Voltage Sag Realities

!
!
!

Magnitude varies during the event

Sag different in Three Phases
Impact of sag event is function of both
voltage magnitude and duration
Impact on equipment is exponential
relationship to voltage
Single index based on integral of voltage
squared and duration Convenient Index
3

Voltage Sag Energy, EVS

VS = [1

Vt

Vnom

dt ] x 100

Sag Energy Index

!

For a given location and period

E
VS _ i
i

System Average Sag Energy

Index

SEI
N

N
6

Voltage Sag Index Definitions

!

SEI Sag Energy Index

!

SCI Sag Count Index

!

Analogous to SAIDI, takes into account voltage sag

frequency, magnitude, and duration.
Analogous to SAIFI, takes into account voltage sag
frequency.

SSI Sag Severity Index

!

Analogous to CAIDI, takes into account voltage sag

magnitude and duration.

SEI = SCI x SSI = month,monitor,etc SE

7

Sample Sag Energy

Calculations
If the phase energy values are 60, 90, and 90 and the duration of the event was 10
cycles (.166 seconds) the Sag Energy (SE) for that event would be:
SE = [(100 60) + (100 90) + (100 90)] x [.166]
SE = [(40) + (10) + (10)] x [.166]
SE = [60] x [.166]
SE = 9.96
SE is only calculated for events up to 15 cycles (.25 seconds) so the worst case
event would be an outage on all three phases for 15 cycles or longer. An
outage is equivalent to phase energy values of 0, 0, and 0. The SE in this case
would be:
SE = [(100 0) + (100 0) + (100 0)] x [.25]
SE = [(100) + (100) + (100)] x [.25]
SE = [300] x [.25]
SE = 75

Sag Event Frequency Index

At a particular location and period
(month or year)
= Number of qualified sag events at that
location and period

System Sag Count Index

Total Number of Qualified voltage Sag events
SCI = -----------------------------------------------------Number of monitor locations
Note: By dividing with number of monitor locations,
data is normalized

10

Qualifying Events
!
!
!
!
!

Voltage Must be < 90%

Aggregation Time: 10 minutes
Event Duration limited to 15 cycles
Energy > 100, = 100
Outages not included

11

Data for 1999 - 2002

1999 2000 2001 2002
No. of Monitors
25
36
43
57
Sag Events (<0.9)
609 926 747 1831
Events after aggregation 408 687 603 1233

12

SEI
(Sag Energy Index)

SSI
(Sag Severity Index)

SCI
(Sag Count Index)

SSCI
(Severe Sag Count Index)

13

SRP Power Quality

Monitoring System, 2002

Sag Energy Index (SEI)

SEI > 35+
25 < SEI < 34
20 < SEI < 24
12 < SEI < 19
SEI < 11

PG
SEI = 31.8

VA (3)
SEI = 29.4

BS (3)

CA
SEI = 44.0

SEI = 9.2

PV (3) [115kV]

PIO (2)

SEI = 9.8

HE (3)

FW (2)

SEI = 10.8

SEI = 39.1

EV (2)
SEI = 14.8

STR
SEI = 6.7

SEI = 13.0

NO (2)

SEI = 7.0

CEM (2)
SEI = 8.3

MIP (2)
SEI = 10.1

WER (2)
SEI = 7.9

MC (2)

PK (3)

SEI = 9.5

CLS

GR

SEI = 8.8

TVH

SR (3)

SEI = 17.3

SEI = 10.3

RO (2)
SEI = 10.2

SEI = 5.7

GL (3)
SEI = 12.8

SEI = 17.0

CE
SEI = 8.2

LE
SEI = 23.7

SS
SEI = 23.0

BL

WA (2)
MEY

SEI = 8.4

SEI = 6.7

GEN (2)
SEI = 7.7

SYY (4)

Installed PQ Monitors (57) *

Future PQ Monitors, FY2002/03 (1)
Includes 3 monitors at PV sub in the Eastern Mining Area
Dashed areas indicate 69kV islands

SEI = 10.3

14

Average SEI (CY2000)

!

Dedicated industrial substation = 12.9

Residential 69kV substation = 18.9

15

16

17

Correlation to C & I Customer

Complaints

L a r g e

In d u s t r ia l

u s t o m

e r

P o lw

e r

u a lit y

Is s u e s ,

Y 2 0 0 0

7 0
6 0
E v e n t

5 0

S e v e r it y

4 0

3 0
2 0
1 0

0
J a n

F e b

a r

p r il

a y

J u n e

J u l y

u g u s t

S e p t

c t

o v

e c

o n t h

18

Sag Energy - FAQs

!

Where do you get sag energy data for

events?
What if PML Monitors are not used?
!

!
!

Product of average sag voltage squared times

duration is a good approximation for integration.
Error < 2 to 3%

Variation of number of monitor locations ?

Data is normalized
Aggregation time ? 1 min or 10 min
SRP LIC Satisfaction Index ?
19

X
Results using Vavg^2/Vnom^2 rather than phase Energy
Results are within +/- 10% towards the low end of the sag energy scale
Maximum deviation at the upper end (longer, more severe events) 20

Conclusions
!

!
!

!
!

SRP installed monitors at selected locations

for benchmarking and quantify power quality
57 Monitors in 2002
Power Quality Index based on sag energy
developed
Good correlation with customers PQ concerns
Results available for four years

21