Post on 28-May-2020
Aidan Tuohy, Jeff Smith and Aminul Huque
July 11, 2014
DG Interconnection Issues and Potential Grid-Support Functionalities
2 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Objectives & Agenda
Agenda:
• Background • Potential Bulk System Reliability Impacts
– Frequency – Voltage – Stability – Other Impacts
• Distribution System Issues
• Cost Considerations
Basic Understanding of Need for Advanced Reliability Services from Inverter-Based Generation
Background
4 © 2014 Electric Power Research Institute, Inc. All rights reserved.
The Electric Power System
5 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Looking Forward
6 © 2014 Electric Power Research Institute, Inc. All rights reserved.
What’s the Big Deal?
US
• 80% of renewable generation impacts distribution
• Vast majority provides no grid support
Germany
• Retrofitting >300,000 solar PV units due to voltage and frequency issues
• €70-180 Million!
“Forward Looking Smart Inverters = A Smart Choice” - Terry Boston, (Summer Seminar, 2013)
“There is an immediate need for new solar to be fitted with “smart inverters” to provide necessary voltage support to integrate effectively and prevent costly renovations and reliability impacts” – Western Electric Industry Leaders, Aug 2013
WEIL Western Electric Industry Leaders
7 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Can Inverter-Based Generation be equipped with
Grid Support Functionality?
Key Question?
8 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Today’s Smart Inverter Technology Enabling Changes in Grid Codes Inverters converts DC energy to AC energy and interface the generating
plant with electricity grid
Traditional Inverter Functionality Smart Inverter Functionality
• Matching plant output with grid voltage and frequency
• Providing safety by providing unintentional islanding protection
• Disconnect from grid based on V/f set points
• Active power reduction in case of over-frequency
• Fault Ride Through (FRT) • Reactive power and voltage
support • Communication with grid • Many more
DC Power AC Power
9 © 2014 Electric Power Research Institute, Inc. All rights reserved.
IEEE 1547 – 2003
• DR shall not actively regulate the voltage at the PCC
• DR shall cease to energize if frequency >60.5Hz
• Tighter abnormal V/F trip limits and clearance times
IEEE 1547a - 2014
• DR may actively participate to regulate the voltage by changes of real and reactive power
• DR shall be permitted to provide modulated power output as a function of frequency
• Much wider optional V/F trip limits and clearance times
• Under mutual agreement between the EPS and DR operators, other static or dynamic frequency and clearing time trip settings shall be permitted.
Major Changes in 1547 Amendment
10 © 2014 Electric Power Research Institute, Inc. All rights reserved.
General Reliability Concerns
Reliability Functions • Reactive power/voltage control • Active power control
– inertia/primary freq response • Disturbance performance
– voltage & frequency ride through
Other Considerations • Inverter capabilities • Available headroom for wind/PV • Distribute
Inverter-Based Generators must supply Reliability Services as they
Displace Conventional
Sources of those reliability services!
11 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Germany: Challenges to New Insights
Source: Fraunhofer Institute, Germany
Recent Changes in Germany to Address Concern of Grid Reliability
Installed Capacity (2013)
Solar
Wind
~64GW of Installed Wind and PV – mostly connected to LV
and MV grid
Coal
Hydro
Nuclear
Gas
Interconnection Rules • Frequency & Voltage Support
Grid Infrastructure Upgrade • ~$27.5B to $42.5B Upgrade
Two Way Communication • Enabled by Advanced
Distribution Management
12 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Don’t Make the Same Mistake!
“50.2 Hz” frequency concern(!) in Germany
4
5
6
7
8
9
10
50 50.25 50.5 50.75 51 51.25 51.5
Inve
rter
Act
ive
Out
put P
ower
(kW
)
Frequency (Hz)
15,000 PV systems Sized 10kW+
9 GW Nominal Power Over 3 – 4 years
Costing €70 – 180 Million
Retrofitting/ Replacing
13 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Inverter – Role in Power System
Frequency Ride Through
Bulk System Issues
15 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Frequency Ride-Through: Risk of Wide-Spread PV Disconnection
50.8
50.6
50.4
50.2
50.0
49.8
49.6
49.4
49.2
Grid
Fre
quen
cy in
Hz
Time in Seconds
0 10 20 30 40 50 60
Source: T. K. Vrana (2011)
Previous Interconnection Rules
German Operating Frequency Limit
Germany
As of 2012 PV Inverters were not required to provide
frequency support and disconnect from the grid if
the frequency reaches 50.2 Hz
This is similar to all current interconnection
requirements in US as per IEEE 1547-2003
16 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Changes in German Grid Code Frequency Support • Frequency control is required of all generators
– Instead of disconnecting @50.2 Hz, gradually reduce active power output (droop curve) in proportion to the frequency
– Retrofit cost mostly through software update ~ $100-250M
Enacted through EEG (German Renewable Energy Sources Act ) & EnWG (German Law on the Energy Industry)
17 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Updated Interconnection Rules Reduces Risk of Frequency Instability
50.8
50.6
50.4
50.2
50.0
49.8
49.6
49.4
49.2
Grid
Fre
quen
cy in
Hz
Time in Seconds
0 10 20 30 40 50 60
Source: T. K. Vrana (2011)
Previous Interconnection Rules
Updated Interconnection Rules German Operating Frequency Limit
Germany
18 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Coordination with Under Frequency Load Shedding – NPCC Directory 12
Source: NPCC Directory 12, p.11, available at: https://www.npcc.org/Standards/Directories/Directory12%20Full%20Member%20Approved%2020130709%20GJD.pdf
19 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Changes in “Response to Abnormal Frequency”
IEEE 1547-2003
IEEE 1547a
Voltage Ride Through
Bulk System Issues
21 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Loss of 250 MW generation
for a normal 230 kV line fault?
Transmission Line Fault
22 © 2014 Electric Power Research Institute, Inc. All rights reserved.
PJM Example LVRT Impact
Source: Mahendra Patel, PJM.
23 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Distributed PV Potential Impact: Low Voltage Ride-Through
FERC Order 661A • Interconnection of wind plants • Requires wind plants remain in
service during – Normally cleared 3-phase
faults to a max. 9 cycles (TOP to provide clearing time)
– 1-phase faults with delayed clearing
– TOP to provide post-fault voltage recovery
Must NOT TRIP Requirement
IEEE Standard 1547-2003** • Interconnection of DER • Requires wind plants disconnect
during
–
Must TRIP Requirement
Voltage Range (% base voltage)
Clearing time(s)
V < 50 0.16 50 ≤ V < 88 2.00
110 < V < 120 1.00 V ≥ 120 0.16
**Proposed Update in 1547a
24 © 2014 Electric Power Research Institute, Inc. All rights reserved.
What is Low Voltage Ride Through (LVRT)?
-600
-400
-200
0
200
400
600
Volta
ge (V
)
-40-30-20-10
010203040
2.05 2.25 2.45 2.65 2.85 3.05 3.25 3.45 3.65
Curr
ent (
A)
Time (s)
6 Cycle Interuption
Inverter Current
Grid Voltage
Inverter able to ride through momentary low voltages and interruptions
25 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Testing L/HVRT Capability EPRI Knoxville Lab
Ride-thru of a voltage sag event to 52% nominal with 15 cycle duration
40%
50%
60%
70%
80%
90%
100%
110%
120%
130%
140%
0.01 0.1 1 10
Volta
ge (%
Nom
inal
)
Time (s)
26 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Low Voltage Ride Through: NERC PRC 024
• Applies to ALL generators • Generator relay requirement • NOT plant ride-through requirement
27 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Voltage Ride-Through: AESO (Canada)
AESO standard specifies via VRT
(high and low) curve, similar to
European approach
28 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Changes in “Response to Abnormal Voltages”
IEEE 1547-2003
IEEE 1547a
29 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Specific VRT Requirements – Germany: MV Grid Code
• Not to disconnect from network in the event of network faults • Support with reactive current • After fault clearance not to extract more inductive reactive power than prior to
fault • Reactive current on LV side of transformer 2% of rated current for each 1% of
voltage dip (outside 10% deadband)
Source: BDEW, Technical Guideline, Generating Plants connected to MV network, 2008)
1500 700 1500 3000 time in ms
U/UC
100%
70%
30%
Below this limit: no need to stay connected
Limit 1: for voltage dips above this limit should not initiate any
disconnection or instatbility Limit 2: voltage dips above this limit (and below limit 1) should
be ridden through
Fault occurance
30 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Voltage Regulation – Reactive Power Support
• Power factor setting
• Constant VAR consumption/injection
• Volt-VAR control
• Voltage Setpoint
31 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Considerations: Reactive Range
• Plant requirements can differ – Reduced range at low
wind/PV levels
• Solar PV inverters over-sized for full range at max power output – historically distribution w/unity
PF control
• Dynamic vs. static capability – switched shunts often
included for static range – SVC/STATCOM may be used
for additional dynamic range
Typical Type 3/4 WTG VAR Capability Options
PV Capability (I limits) Vs. Triangle Require.
Frequency and Voltage Stability
Bulk System Issues
33 © 2014 Electric Power Research Institute, Inc. All rights reserved.
High Levels of Inverter-Based Generation Can Impact Frequency Stability
Graphics Source: LBNL-4142E Use of Frequency Response Metrics to Assess the Planning and Operating Requirements for Reliable Integration of Variable Renewable Generation, Prepared for Office of Electric Reliability Federal Energy Regulatory Commission, Dec 2010
Sudden Generation
Loss Non-synchronous VG may provide inertia & primary
frequency response through power
electronics
VG may displace conventional units &
reduce system inertia & primary
frequency response
34 © 2014 Electric Power Research Institute, Inc. All rights reserved.
59.5
59.6
59.7
59.8
59.9
60
60.1
0 5 10 15 20 25 30 35 40 45
Freq
uenc
y (H
z)
Base(15%)
Base (20%)
Base(30%)
Base(40%)
Nadir
Settled Frequency
UFLS
EPRI Frequency Response Project (WECC) Impact of Wind Without Frequency Response
All Else Constant, as Wind Gen without Freq Control
replaces Thermal Gen, Freq Performance Deteriorates
Similar results likely with DER
35 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Example EPRI Simulation Results: Generic Case
As PV Increases Voltage Recovery Time Increases
36 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Simulation Studies : Steady State Assessment
Identify Critical Contingencies and Buses •Ranking •Maximum Voltages Deviations •Lowest Voltages
Perform PV analysis •Determine real power transfer or load level limit
•Test for N,N-1-1,N-1 and N-2 conditions
Perform QV analysis •Determine reactive power margins at critical buses
•Test for N,N-1 and N-1-1 conditions
37 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Simulation Studies : Dynamic Assessment
Impact on Voltages :
Delay in voltage recovery
With and without voltage ride-thru
Impact on Frequency:
Reduction in system inertia due to change in generation mix
PV drop out due to large voltage disturbance (as per IEEE 1547)
Other potential issues
Bulk System Issues
39 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Germany’s System Curve with Increasing PV
40 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Real power control, ramping, and curtailment
100% 75% 25% In 4 seconds In 6 seconds
41 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Active power control / frequency support
• Germany: 10% of grid connection capacity per minute – Above 50.2 HZ, active power reduces at 40% per Hz
• Ireland, ramp rate 1-30MW/min – Active power depends on frequency:
• Above 52 Hz, no active power, • 50.5 -52 Hz, power goes from 100-20% active power
• Nordic: 10% of rated power per minute • Denmark: Active power depends on frequency • GB: Frequency control device for primary and secondary
frequency control, plus over frequency control
42 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Communications with DER • Possibility for DSO to send setpoint values to generators
– Reduce the active power output or change the cos(phi). – Guarantee grid stability in case of emergency situations or congestion
• Can be implemented in various ways: – Audio or radio frequency ripple control – IEC 61850-90-7 – IEC 60870-5-104
Source: R. Bründlinger, Austrian Institute of Technology, “Grid Code Experiences from Europe”, EPRI webcast, May 2014
43 © 2014 Electric Power Research Institute, Inc. All rights reserved.
What’s EPRI Doing in Distributed Resource Integration in Bulk System
Ongoing and Near-Term EPRI Efforts Project/Effort Topic(s) Addressed Timeline
Voltage and Frequency Performance with High Levels of VG
•Impacts of distributed PV on bulk system performance •Grid code/ Requirements •Impact on frequency response and new resources to provide FR
Ongoing
DOE SUNRISE project • TVA & Southern Company strategic plans for PV impacts on dist and bulk system, business models and IRP •CAISO/SMUD production simulations • Coordinate w/ projects in P173 and P174
2013-2015
EPRI White Paper on Demand Response
• Completed end 2013 2013
EPRI Value of the Grid Concept Paper
• Begun end 2013 2013+
Distribution Impacts
45 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Distribution System Impacts
Voltage • Overvoltage • Voltage variations
Equipment Operation • Feeder regulators • Load tap changers • Switched capacitor banks
Demand • “Masking” peak demand • Reducing power factor
System Protection • Relay desensitization • Unintentional islanding
Power Quality • Harmonics
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
7 8 9 10 11 12 13 14 15 16 17 18 19
Pow
er (p
er-u
nit)
Hour
46 © 2014 Electric Power Research Institute, Inc. All rights reserved.
= Line Regulators= Substation
0.95
0.97
0.99
1.01
1.03
1.05
1.07
1.09
0 60 120 180 240 300 360 420 480 540
Volta
ge (p
u)
Time (sec)
0.95
0.97
0.99
1.01
1.03
1.05
1.07
1.09
0 60 120 180 240 300 360 420 480 540
Volta
ge (p
u)
Time (sec)
Why are Smart Inverters Important? Smart Inverters Mitigating Voltage Issues
46
Voltage at END of feeder
No Control Volt/Var Control
*Simulated in OpenDSS
47 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Use of Smart Inverters for Accommodating Large Number of Distributed PV Systems
24 Hour Simulation
Source: J. Smith, T. Key “High-Penetration PV Impact Analysis on Distribution Systems,” Solar Power International, Oct 2011
Customer Load Customer PV
Primary Voltage
0.9
0.925
0.95
0.975
1
1.025
1.05
0 4 8 12 16 20
Hour
Vol
tage
(pu)
Baseline – No PV
20% PV
Solar Rooftop PV
48 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Use of Smart Inverters for Accommodating Large Number of Distributed PV Systems
24 Hour Simulation
Source: J. Smith, T. Key “High-Penetration PV Impact Analysis on Distribution Systems,” Solar Power International, Oct 2011
VARs
Gen
erat
ed
Capacitive
Inductive
System Voltage
V1 V2 V3 V4
Q1
Q4
Q3 Q2
Primary Voltage
0.9
0.925
0.95
0.975
1
1.025
1.05
0 4 8 12 16 20
Hour
Vol
tage
(pu)
Baseline – No PV
20% PV 20% PV with
volt/var control
Initial analysis indicated 25%-100% more PV can be accommodated using
Volt/var control
Solar Rooftop PV
With volt/var control Volt-Var Control
49 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Increased Hosting Capacity using Smart Inverters
PV at Unity Power Factor PV with Volt/var Control
2500 cases shown Each point = highest primary voltage
ANSI voltage limit
ANSI voltage limit
Increasing penetration (kW)
Max
imum
Fee
der V
olta
ge (p
u)
Max
imum
Fee
der V
olta
ges
(pu)
Increasing penetration (kW)
No observable violations regardless of PV size/location
Possible violations based upon PV size/location
Observable violations occur regardless of size/location
For voltage-constrained feeders, results indicate use of smart inverters can increase feeder hosting capacity for PV
Minimum Hosting Capacity Maximum Hosting Capacity
Minimum Hosting Capacity Max Hosting Capacity
50 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Ongoing Research – Smart Inverter Settings
Objective Determine recommended default settings for actual field deployments of smart inverters
Approach
Time-series simulations in OpenDSS comparing feeder performance with and without smart inverter functions
Value Provide utilities with guidelines for actual PV system installations
Sites Model-based analysis of 3-phase PV systems along with field demonstrations
Wide range of possible smart inverter settings*
*select few shown for illustration
51 © 2014 Electric Power Research Institute, Inc. All rights reserved.
What Smart Inverter Settings are “Best?”
0 5 10 15 20 25
1.024
1.026
1.028
1.03
1.032
1.034
1.036
1.038
1.04
1.042
1.044
Hour
Vol
tage
(pu)
g g
---- Voltvar
---- No PV---- PV base
Daily Distribution Feeder Voltage Profile with different volt/var setpoints
52 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Summary – Smart Inverter Settings
• Overall “best” setting depends upon objective – Improve voltage – Increase efficiency – Decrease regulator operations – Increase hosting capacity
• Preliminary analysis indicates trends in recommended settings can be found
• Caution: Minor changes in settings (volt/var) can have significantly different impacts
• Less “aggressive” settings work – Less risk, less potential benefit
(e.g., increasing hosting capacity)
• Future work for determining recommended settings – Other locations – Other feeders – Combined inverters
53 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Ongoing Research - Smart Inverter Interaction
Objective Evaluate Potential Inverter Interaction Approach
Investigate possible inverter interaction resulting from smart inverter control on multiple PV systems through modeling
Var control flow
VoltVar CurveInverter
Averagingwindow
VReference Q
Average V
QPI
controller
(Kp Ki)
Switching Command
Factors may impact var control:• Volt-var curve parameters• PI controller parameters: Kp and Ki
• Voltage average window length
54 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Initial Results
• Interactions exist between the two close by inverters
• High ratio may cause oscillations
• Smaller adjustments at each step as a result of smaller control parameters reduce oscillations
• Longer voltage averaging window increases magnitude of oscillations, although dampening does occur
• True test will be field demonstrations
4 6 8 10 12 14 16 18 200.95
0.96
0.97
0.98
0.991
1.01
1.02
1.03
1.04
1.05
Time(s)
Vol
tage
(pu)
4 6 8 10 12 14 16 18 200.950.96
0.97
0.980.99
1
1.011.02
1.03
1.041.05
Time(s)
Vol
tage
(pu)
Single PV System
2 PV Systems
Oscillations Observed
Cost Considerations
56 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Inverter Cost and Reliability Impacts PV System Performance
• Inverter cost is a small percentage of the total plant cost/installed cost
• Inverters with grid support functions are not expected cost any higher than current price once the interconnection requirements start incorporating the functional capabilities
• Communication (where needed!) cost will be additional
Average U.S. Utility-Scale PV Price Breakdown (May 2013)
Q1 2013 US PV System Cost Residential - $4.93/W
Commercial-scale - $3.92/W Utility-scale - $2.14/W
National average - $3.37/W
Sources: Sources: GTM Research/SEIA, NREL, Lawrence Berkeley National Lab, BNEF, European Photovoltaic Industry Assn (EPIA), BSW Solar
57 © 2014 Electric Power Research Institute, Inc. All rights reserved.
0
25
50
75
100
125
150
175
200
225
2005 2010 2015 2020 2025 2030 Inst
alle
d PV
Cap
acity
(GW
) in
USA
Years
What if Standards are not Changed?
Source: DOE SunShot Vision Study (Total PV Capacity in U.S. 2008 – 2030)
Retrofitting inverters in a later date could be pretty expensive
58 © 2014 Electric Power Research Institute, Inc. All rights reserved.
Together…Shaping the Future of Electricity