Using Full-Time Power Quality
Monitoring as an Asset
Management Tool.
• EDIST 2014 Conference
• January 16, 2014
• Toronto Ontario, Canada
• Presented by: Gary MacLeod,
•
Project Manager, Power Quality Enhancement Program
Privileged and Confidential
1
Objectives…
• Demonstrate how Hydro Ottawa uses their Full-Time
PQ Monitoring System as an Asset Management Tool
– Detect defective equipment before catastrophic failure
– Verifying Protection and Control settings
– Fault location
– Provide better customer service
– Reduce customer PQ investigations
– Benchmarking PQ indices before installation of DG or new LRT
Privileged and Confidential
2
Hydro Ottawa Power Quality Monitoring
• Full-time Power Quality monitoring Using Electrotek Concept's
“PQView” software and Schneider ION meters and ION
Enterprise software
• We started in 2002 with a small numbers of ION 8500 meters at
some of our IESO Revenue Meter Points.
• We currently have 123 PQ meters installed or planned for this
year and 10 more planned to monitor all buses except 4.16kV
• Current project underway to acquire PQ data and waveform
events from our SEL relays and import it into PQView
• Current project underway to use PQView and CYME for fault
locating
Privileged and Confidential
3
Hydro Ottawa Power Quality Monitoring
Revenue PQ Meters
End Users
ION Enterprise
PQView Database
PQ Meters
PI Historian
Privileged and Confidential
CYME Database
4
Typical Metering Single-line Drawing
Main
Revenue/PQ
Meter
Privileged and Confidential
5
History of Recent Tapchanger Failures
• We had three on-load tap-changers fail in past 36
months
• Transformer 1 was a 115kV – 16/27.6kV 33 MVA
• Transformer 2 was a 115kV – 16/27.6kV 37 MVA
• Transformer 3 was a 115kV – 4.8/8.3kV 6 MVA
• Different transformer manufacturers
• Two different tapchangers
Privileged and Confidential
6
Waveforms from 1st defective tapchanger
Transformer 1 - 21/04/2010 06:13:41.1825
Vb
Vc
Ia
Ib
Ic
Va
100
50
50
Voltage (%)
100
0
-50
200
0
-200
-400
Vc
Ia
Ib
Ic
0
-100
0
Vb
-50
-100
400
Current (A)
Current (A)
Voltage (%)
Va
Transformer 1 - 22/04/2010 02:04:12.5558
-500
-1000
0
0.005
0.010
0.015
Time (s)
0.020
0.025
0.030
Va
Vb
Vc
Ia
0
0.02
0.04
0.06
0.08
0.10
0.12
Time (s)
Transformer 1 - 25/04/2010 01:51:28.2487
Hydro Ottawa - Technical Services
PQView®
Ib
Transformer 1 - 25/04/2010 11:50:35.5390
Hydro Ottawa - Technical Services
Ic
Va
Vb
Vc
Ia
Ib
PQView®
Ic
100
100
Voltage (%)
Voltage (%)
50
0
-50
-100
0
-50
-100
50
0
200
-50
Current (A)
Current (A)
50
-100
-150
100
0
-200
-250
0
0.05
Hydro Ottawa - Technical Services
0.10
0.15
Time (s)
0.20
0.25
0.30
PQView®
-0.02
0
Hydro Ottawa - Technical Services
Privileged and Confidential
0.02
0.04
Time (s)
0.06
0.08
0.10
PQView®
7
Transformer 1 RMS Variations and Damage
RMS Variation Magnitude Time Scatter Plot
Transformer 1
Voltage Magnitude (pu)
0.8
0.6
0.4
0.2
Timeline shows that the
magnitude of the voltage
sags quickly got worse
before transformer was
taken out of service
0
8 Thu
Apr 2010
Hydro Ottawa - Technical Services
15 Thu
Time
22 Thu
PQView®
Privileged and Confidential
8
Transformer 3 - V RMS AB, V RMS BC, V RMS CA
From 31/03/2010 12:00:00 PM to 15/01/2012
Transformer 3 “No Events”
110
• There were no
waveshape events
recorded during this
period.
• Voltage Trend data
showed a frequent
number of red phase
minimum values that
were not associated with
current inrush or faults.
• Transformer was taken
out of service and
inspected and damage
was found.
Min[V RMS AB] (%)
Avg[V RMS AB] (%)
Max[V RMS AB] (%)
Min[V RMS BC] (%)
Avg[V RMS BC] (%)
Max[V RMS BC] (%)
Min[V RMS CA] (%)
Avg[V RMS CA] (%)
Max[V RMS CA] (%)
105
100
95
Transformer 3 - I RMS A, I RMS B, I RMS C
90
From 31/03/2010 12:00:00 PM to 15/01/2012
2 Mon
3 Tue
4 Wed
5 Thu
Min[I
RMS
A]
(A)
Avg[I
RMS
A]
(A)
Max[I
RMS
A]
(A)
Min[I RMS B] (A)
Jan 2012
Time
Hydro Ottawa
- Technical
Avg[I
RMS B] (A) Services
Max[I RMS B] (A)
Min[I RMS C] (A)
PQView®
Avg[I RMS C] (A)
Max[I RMS C] (A)
300
200
100
0
1 Sun
Jan 2012
2 Mon
Hydro Ottawa - Technical Services
Privileged and Confidential
3 Tue
4 Wed
5 Thu
6 Fri
Time
PQView®
9
Voltage PLT Flicker Trends
Transformer 1 - Plt A, Plt B, Plt C
Transformer 2 - Plt A, Plt B, Plt C
From 31/03/2010 12:00:00 PM to 15/01/2012
From 31/03/2010 12:00:00 PM to 15/01/2012
Plt A
Plt B
Plt C
Plt A
3.0
Plt B
Plt C
8
2.5
7
6
2.0
5
1.5
4
3
1.0
2
0.5
1
0
0
15 Thu
22 Thu
Apr 2010
8 Mon
Time
Hydro Ottawa - Technical Services
Aug 2011
PQView®
Transformer 3 - Plt A, Plt B, Plt C
Hydro Ottawa - Technical Services
22 Mon
PQView®
Transformer 2
From 31/03/2010 12:00:00 PM to 15/01/2012
Transformer 1
Plt A
15 Mon
Time
Plt B
Plt C
1.4
1.2
PST and PLT flicker data identified
the start of the serious damage to
the tapchanger within hours on all
three transformers.
1.0
0.8
0.6
0.4
0.2
0
15 Thu
Dec 2011
22 Thu
1 Sun
8 Sun
15 Sun
Time
Hydro Ottawa - Technical Services
PQView®
Transformer 3
Privileged and Confidential
10
Failed Operation of New Capacitor Bank
HO_Ellwood.T1e - 15/01/2013 11:37:41.1145
Duration: 9.437s
Va
Vb
Vc
Ia
Ib
Ic
Voltage (V)
10000
0
-10000
Current (A)
500
0
-500
0
1
2
3
4
5
Time (s)
6
Event ID = 288312
Electrotek/EPRI
7
8
9
10
System Event ID = 51733
PQView®
This new capacitor bank switched in and out seventy times in ten
seconds. Engineering staff and control manufacturer believed this
could not happen and the meter was erroneous.
Privileged and Confidential
11
Failed Operation of New Capacitor Bank Confirmed
HO_Ellwood.T2e - 15/01/2013 11:37:38.9792
Duration: 9.437s
Va
Vb
Vc
Ia
Ib
Ic
In this substation, both buses are in
parallel and both ION meters saw
the same event and SCADA told us
the breaker operated a number of
times
Voltage (%)
100
0
-100
Current (A)
500
0
HO_Ellwood.T2e - 15/01/2013 11:37:38.9792
-500
Total Power Factor
ΔPFa=-0.001469 PF ΔPFb=0.0006440 PF ΔPFc=-0.0004120 PF ΔPFtotal=-0.0004854 PF
1.55
PFtotal
1.60
1.65
1.08
Event ID = 288314
Electrotek/EPRI
1.06
1.70
1.75
1.80
Time (s)
1.85
1.90
System Event ID = 51733
1.95
2.00
Reference Cycle = 738056
PQView®
1.04
Total Power Factor (pu)
The graph below shows the power
factor is fluctuating between events
and leading for a few cycles after
the event and then dropping back to
0.96
1.02
Our engineering staff and the
manufacturer are now investigating
further to determine the cause
1.00
0.98
0.96
0.94
0.92
0.90
0.88
0.86
1.55
Electrotek/EPRI
1.60
1.65
1.70
1.75
1.80
Time (s)
1.85
1.90
1.95
Privileged and Confidential
2.00
PQView®
12
Transformer Voltage Imbalance Problem
HO_Bridlewood.T1e - V Imb Total
From 01/01/2013 to 14/10/2013
2.5
In June last year, a substation ION started
to indicate increased voltage imbalance on
a transformer bus.
Avg[V Imb Total] (%)
2.0
The phase-to-phase voltage graph below
did not indicate anything significantly
abnormal
1.5
1.0
0.5
0
13 Thu
Jun 2013
14 Fri
15 Sat
16 Sun
17 Mon
18 Tue
19 Wed
20 Thu
HO_Bridlewood.T1e - V RMS AB,
V RMS BC, V RMS CA
Time
21 Fri
From 01/01/2013 to 14/10/2013
Electrotek/EPRI
Avg[V RMS AB] (%)
Avg[V RMS BC] (%)
Problem was a broken chain on one phase
in a tap-changer only allowing two phases
to be regulated and the over fluctuated with
the 115kV supply
22 Sat
PQView®
Avg[V RMS CA] (%)
106
Recommendation: Setup 3 phase- 4 wire
wye ION meters to measure phase-toneutral voltages rather than phase-tophase. This can be done after a meter is
locked and sealed as well.
105
104
103
102
101
100
99
98
15 Sat
Jun 2013
Electrotek/EPRI
22 Sat
Time
1 Mon
PQView®
Privileged and Confidential
13
Integrated with SCADA PI Historian to correlate
breaker events with PQView event captures
Sag event
captured by
ION meter
Breaker
Operation
imported from
SCADA
Event Cause added and available to
all users and characteristics added
for future analysis
Privileged and Confidential
14
Using PQView and CYME for Fault Locating
HO_Overbrooke.T1_J2 - 10/7/2013 11:49:38.6471
Duration: 1. 099s
Va
Sag event captured
by ION meter
Voltage (%)
TO_TO30_CB
10/7/2013 11:49:57.0000
Overbrook TO30 Breaker Status
CLOSE-OPEN
Breaker Operation
imported from SCADA
Reactance to Fault
measurement
calculated by PQView
Vc
Ia
Ib
Ic
100
50
0
-50
-100
5000
Curre nt (A)
Operation
Point Name
Time Stamp
Value
Description
Vb
0
-5000
1C|Rs 0=0. 3551|X s0= 3.717
0
0.5
1.0
1.5
Time (s)
Hydro Ottawa Technical Services
PQView®
Privileged and Confidential
15
Using PQView and CYME for Fault Locating
Reactance to
Fault
measurement
calculated by
PQView
PQView
determines
possible locations
of fault on
CYMDIST circuit
model
Privileged and Confidential
16
Using PQView and CYME for Fault Locating
As Shown on Google Earth
Orange line is UG cable
Predicted fault locations
Yellow line is overhead section
Substation
ACTUAL FAULT LOCATION
Predicted fault location
Privileged and Confidential
17
PQView Trend Data Analysis
Trend Chart with Hyperlinks to PQ Events
Profile Charts (this example is hourly)
Histogram Chart with Statistics
Privileged and Confidential
18
Sag/Swell Analysis (5-Year period)
Events plotted on ITIC Chart
SARFI Event Count
Events plotted on Time Scatter graph
Sag Magnitude Column Chart
Privileged and Confidential
19
“Big Data”
Average 3-phase L-L Voltage
Meter Location
Meter Site 1
Meter Site 2
Meter Site 3
Meter Site 4
Meter Site 5
Meter Site 6
Meter Site 7
Meter Site 8
Meter Site 9
Meter Site 10
Meter Site 11
Meter Site 12
Min
CP00.5
<90
<90
73.99
98.5
98.57
99.21
99.61
100.7
97.41
99.39
99.4
99.48
99.68
100.9
100
100.7
100
100.7
97.52
98.56
101.4
101.6
101.2
101.2
97.11
98.09
CP01
100.8
99.38
100.9
99.79
99.51
101
100.8
100.8
98.87
101.7
101.3
98.29
CP05
<95
101.4
99.98
101.5
100.8
99.7
101.9
101.8
101.8
99.34
101.9
101.5
98.96
CP25
102.1
101.1
102.2
101.6
100.1
102.3
102.3
102.3
100.2
102.3
101.9
99.74
Avg
102.5
101.9
102.8
102.1
100.5
102.6
102.6
102.6
100.7
102.7
102.3
100.3
CP50
102.6
101.8
102.7
102.2
100.5
102.6
102.6
102.6
100.8
102.7
102.3
100.3
CP75
103.1
102.7
103.3
102.7
100.9
103
102.9
102.9
101.3
103
102.7
100.9
CP95
>105
103.9
103.8
104.2
103.5
101.2
103.4
103.3
103.3
101.8
103.5
103.1
101.9
CP99
104.2
104.5
104.8
103.9
101.4
104.1
103.6
103.6
102.1
103.7
103.3
102.5
CP99.5
104.4
104.7
105.1
104
101.4
104.3
103.9
103.9
102.2
103.8
103.4
102.8
Max
>110
104.6
105.3
105.4
104.3
101.6
104.9
104.8
104.8
102.9
104
103.6
103.3
Count
σ
2880
2880
2880
2880
2880
2880
2880
2880
2880
2218
2880
2880
2.109
1.139
0.8251
0.8596
0.4836
0.5235
0.5015
0.5015
0.7585
0.4652
0.5061
0.8848
Databases now contain large amounts of data and the challenge is to be able to use
the data to find and repair trouble spots quickly.
This spreadsheet shows 12 sample sites with monthly line-to-line voltage statistics for
November 2013. The cells are condition formatted and quickly highlight trouble spots
that should be reviewed.
Meter Site 1 is a rural wholesale meter point that experienced an extended voltage sag
when an upstream single-phase recloser locked out during a windstorm. The stats
show that the remainder of the voltage measurements were within predetermined
thresholds
Privileged and Confidential
20
Using Big Data…
Fault locations
shaded by age.
Brightest are the
most recent
Recent lightning
strike correlates with
known fault location
and auto-relocate
Rash of events in
same area
Trees?
Bad insulators?
Large population of
squirrels?
Privileged and Confidential
21
PQ Data Use and Future Potential
Basic Functionality Now
Future Potential
•Relay Integration
•Outage Management System
(OMS) Integration
•Automated Reporting
•Standards Compliance
•Predictive Maintenance
•Fault Locating
•Event Alert
Privileged and Confidential
22
Did we cover everything?
– Detect defective equipment before catastrophic
failure
– Verifying Protection and Control settings
– Fault location
– Provide better customer service
– Reduce customer PQ investigations
– Benchmarking PQ indices before installation of DG
or new LRT
Privileged and Confidential
23
Distribution Operations
Top 2013 Projects
Questions?
Contact:
Gary MacLeod, Project Manager, Power Quality Enhancement Program
Tel: 613 738-5499 ext 7108
garymacleod@hydroottawa.com
Privileged and Confidential
24