)
Islanding
Voltage Regulation
Power Quality
Reclosing Coordination
Sectionalizing
Fuse Coordination
Short Circuit Contribution
Capacitor Switching
Stability
Adjacent Feeder Faults
Wide and growing range of both inverter-based and rotating
machine-based technologies, all selected based on current and
potential market impact
Includes three classes of equipment (as of 6/04)
Residential (single-phase, 5 kW)
Commercial (three-phase, 30 kW to 250 kW)
Industrial (three-phase, 250 kW) June 21, 2004
3
Distributed Utility Associates
Single Line Diagram
Bay 1
2.5
2.5
2.5
2.5 June 21, 2004
4
Distributed Utility Associates
Single Line Diagrams
Bay 2 and 3 June 21, 2004
5
Distributed Utility Associates
The DUIT Facility
Bay 1 Residential DR
Distributed Resources
2.5 kW, 240 VAC, 1-ph PV Inverters (10)
4.0 kW, 120 VAC, 1-ph Battery-Based Inverters (10)
2.5 kW, 240 VAC, 1-ph PV Inverters (8)
5.0 kW, 120 VAC, 1-ph Battery-Based Inverters (6)
Power Sources
12.5 kW PV Array
Various power supplies rated from 60VDC to
600VDC at power ratings to 6 kW
Load Banks
120V & 240V Type Test load banks,
5 kW/±15 kVAR each, independent
variac control on R, L and C
Three 25 kW/±75 kVAR load banks
for multi-unit testing June 21, 2004
6
Distributed Utility Associates
The DUIT Facility
Bay 2 Commercial DR
Distributed Resources
100 kW, 480V 3-ph PV Inverter
45 kW, 480V 3-ph PV Inverter
Power Sources
PV Simulator (Power Supply)
300 to 600VDC, 2x300ADC
Load Bank
400 kW of resistive load, 750 kVAR each of inductive and
capacitive load
Coarse control (DUIT):-Facility-and-Unintentional-/' >through discrete steps of R, L, C
Fine control (DUIT):-Facility-and-Unintentional-/' >through variacs on 25 kW of resistance, 25 kVAR of
capacitance June 21, 2004
7
Distributed Utility Associates
The DUIT Facility
Bay 3 Industrial DR
Distributed Resource
240 kW, 300 kVA, 480V 3-phase diesel engine-driven
synchronous generator
Load Bank
Shares 400 kW and
± 750 kVAR load
with Bay 2 June 21, 2004
8
Distributed Utility Associates
The DUIT Facility
Medium Voltage Distribution System
Each bay tied into PG&E
medium voltage distribution
system via one or more pole-
mount transformers
21 kV motor-operated load
break switch allows for isolation
of total DUIT facility from grid June 21, 2004
9
Distributed Utility Associates
The DUIT Facility
Data Acquisition and Control System
National Instruments LabView based system
6 kHz sampled data collected for voltage and current
at every relevant node in system for each test June 21, 2004
10
Distributed Utility Associates
Anti-Islanding Test
Plan
Test Description and Sequence
6.1
Basic Islanding Test
Individual unit testing
6.2
Islanding with Multiple DRs
Homogeneous Groups
Small Groups
Progressions
6.3
Non-Linear Loads, Anti-islanding Tests
Individual Units
6.4
Islanding with Dynamic Load:Generation Ratios
Individual Units
6.5
Anti-islanding with Rotating Loads
Individual Units
6.6
Harmonic Content due to Anti-islanding Schemes
Individual Units
6.7
Voltage/Frequency Trip Settings
Individual Units
Homogeneous Groups
Small Groups
Progressions June 21, 2004
11
Distributed Utility Associates
Single Unit Islanding Test Procedure
Basic anti-islanding tests (DUIT Test Protocol 6.1)
performed on a single unit for each DR model (two
single units for single phase inverters)
IEEE 1547/UL 1741/CA Rule 21 procedures
Main purposes:
Ring out test setup and data acquisition in preparation
for multi-unit islanding tests
Ensure test setup produces consistent results from unit to
unit
Gain an understanding of manufacturers anti-islanding
methods to select appropriate combinations for multi-unit
testing June 21, 2004
12
Distributed Utility Associates
Single Unit Islanding Test Procedure
Measure DR shutdown times under three generation/
load conditions
P
gen
= 25%, P
load
= 25%
P
gen
= 50%, P
load
= 50%
P
gen
= 100%, P
load
= 100%
In each case, inductive and capacitive loads are
adjusted to resonant frequency of 60 Hz with quality
factor (Q) of 2.5 for first test
L or C is adjusted in steps of 1% between -5% and
+5% of nominal value used in first test
Results in a total of 33 tests per DR June 21, 2004
13
Distributed Utility Associates
Single Phase Test Setup
Power Supply
0-60/600VDC
0-100/10ADC
Inverter
21 kV
21kV:240/120V
Island
Contactor
120/240V Load Bank
0 - 5 kW
0 - 15 kVAR Ind
0 - 15 kVAR Cap June 21, 2004
14
Distributed Utility Associates
Three Phase Test Setup
480 V
Inverter
Inverter
100 kVA
208 V:480 V
45 kVA
208 V:480 V
PV Simulator
0-600VDC
0-300ADC
450 kW/±750 kVAR 3-ph 480V load bank
21kV
150 kVA
21 kV:480 V
Island Contactor
21kV
150 kVA
21 kV:480 V
Island Contactor
G
MO
240 kW/300 kVA
Diesel Engine
Generator
June 21, 2004
16
Distributed Utility Associates
Single Unit Islanding Test Results
Bay 2, Initial Testing
Device
Mean
Trip Time
Median
Trip Time
Standard
Deviation
Trip Times
> 2
seconds
Longest
Trip Time
Device E
0.12 s
0.10 s
0.05 s
0 of 33
0 of 33
0.39 s
Device F
0.26 s
0.20 s
0.23 s
1.16 s June 21, 2004
17
Distributed Utility Associates
Single Unit Islanding Test Results
Bay 3, Initial Testing
Device
Mean
Trip Time
Median
Trip Time
Standard
Deviation
Trip Times
> 2
seconds
Longest
Trip Time
Device G*
1.27 s
0.99 s
0.66 s
5 of 40
3.33 s
* This device includes external anti-islanding controls that have not yet been NRTL-
certified to UL. June 21, 2004
18
Distributed Utility Associates
Single Unit Islanding Test Results
Bay 1, Initial Testing
Device
Mean
Trip Time
Median
Trip Time
Standard
Deviation
Trip Times
> 2
seconds
Longest
Trip
Time
A
0.34 s
0.34s
0.08 s
0 of 66
1 of 66
0 of 66
1 of 66
0.55 s
3.31 s
1.16 s
2.22 s
B
0.27 s
0.09 s
0.48 s
C
0.14 s
0.06 s
0.23 s
D
0.20 s
0.08 s
0.35 s June 21, 2004
19
Distributed Utility Associates
Single Unit Islanding Test Results
Bay 1, Supplementary Testing
Device B
Longest run-on in initial testing occurred at rated power; inverter
operating in power limiting mode
Supplemental testing with device at power limit and DC input
voltage near minimum of operating range resulted in some trip
times exceeding 10 seconds with well-tuned load
Device C
When power levels were decreased from 27% to 24% of device
rating, run-ons in excess of 30 seconds were observed.
Device D
Power limiting in initial testing achieved (DUIT):-Facility-and-Unintentional-/' >through changing inverter
setpoints, consistent with UL methods
Supplemental testing performed by lowering input voltage to effect
DC current limiting for achieving partial load points; some run-on
times in excess of 30 seconds were observed with well-tuned load June 21, 2004
20
Distributed Utility Associates
Single Unit Islanding Test Results
Resulting Follow-On Actions
Recommended revisions to UL1741:
1) Manufacturer to provide details of anti-
islanding algorithm, or 2) device to be subjected
to islanding tests in center and limits of input
voltage window
5% load (quadruples # of test points)
More explicit language regarding:
Definition of balanced test conditions
Test load linearity
Tolerances on pre-test voltage and frequency
Accounting for parasitic losses (real and reactive) June 21, 2004
21
Distributed Utility Associates
Anti-Islanding Test
Plan
Test Description and Sequence
6.1
Basic Islanding Test
Individual unit testing
6.2
Islanding with Multiple DRs
Homogeneous Groups
Small Groups
Progressions
6.3
Non-Linear Loads, Anti-islanding Tests
Individual Units
6.4
Islanding with Dynamic Load:Generation Ratios
Individual Units
6.5
Anti-islanding with Rotating Loads
Individual Units
6.6
Harmonic Content due to Anti-islanding Schemes
Individual Units
6.7
Voltage/Frequency Trip Settings
Individual Units
Homogeneous Groups
Small Groups
Progressions
Done
In Prog
ress June 21, 2004
22
Distributed Utility Associates
Distribution System Impact Study
Initial study performed to look at the probability of
necessary conditions for an islanding event to occur
I Static P, Q balance
0.001 to 0.01 per event (High penetration)
1E-10 to 1E-8 per event (Low penetration)
II Resonant Frequency
1E-5 to 1E-2 per year
III Fault events
0.15 to 0.3 per year
IV Time domain stability
0.1 per year
All conditions are statistically independent
High penetration: 1.5E-10 to 3E-6 islands per feeder per year
Low penetration:1.5E-17 to 3E-12 islands per feeder per year
Note: Study assumes over/under voltage and frequency protection only, i.e. no active anti-islanding schemes