Webinar 4 - Aggregated Slides
Webinar 4 - Aggregated Slides
Webinar 4 - Aggregated Slides
2
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Motivation, Challenges & Needs
• Continuously increasing penetration level of inverter
interfaced resources, predominantly renewables (Type
III, Type IV WTGs & PVs)
Challenge
4
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Wind/PV Short Circuit Models
5
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Inverter Based Resources Short-Circuit Modeling
6
© 2018 Electric Power Research Institute, Inc. All rights reserved.
EPRI Phasor Domain Model for SC Calculations
Type IV WTG Type III WTG Solar PV
Wind Wind
turbine Stator power
turbine Grid
Stator-Side Grid-Side Gearbox
Slip rings
Gearbox Converter Converter IL , P L Transformer
iPMSG ig Chopper
Grid
Rotor power
Step down Crowbar
Rotor -Side Grid -Side
transformer Converter Converter
WTG (Type III & Type IV) and Solar/PV phasor domain short circuit model:
• Voltage controlled current source
• Iterative solution (nonlinear behavior)
• considers the impact of controls (reactive power/voltage control) on the short circuit response
• respects converter current limits
7
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Inverter Control Mode Options
Function Control Mode Performance/Description
Constant power factor Allows for inverter injection/absorption of
reactive power based on a desired power
factor
Constant Q Allows for inverter fixed desired value of
Reactive power/voltage
reactive power injection/absorption
control during ride-
through V Control Allows for inverter control of voltage to desired
value
Dynamic reactive current control based on Allows for reactive current injection based on a
reference curve reference curve (e.g. grid code)
8
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Current Limiter - PQ Priority - Examples
Example 1:
Assume: Desired Currents: Upon current limiter:
Active Power: 1 p.u. Iactive= 1/0.7=1.43 p.u Iactive= 0.92 (reduced to
Post fault voltage: 0.7 pu satisfy limit)
Ireactive=2(1-0.7) = 0.6 p.u
Control mode: Reference curve with slope 2 Ireactive= 0.6 p.u
Itotal=1.55 pu (exceeds
Q priority Itotal= 1.1 pu
limit)
Ilimit=1.1 pu
Example 2:
Upon current limiter:
Assume: Desired Currents:
Iactive= 0 (reduced to satisfy
Active Power: 1 p.u. Iactive= 1/0.4=2.5 p.u
limit)
Post fault voltage: 0.4 pu Ireactive=2(1-0.4) = 1.2 p.u
Ireactive= 1.1 p.u (reduced
Control mode: Reference curve with slope 2
Itotal=2.77 pu (exceeds to satisfy limit)
Q priority limit)
Itotal= 1.1 pu
Ilimit=1.1 pu
9
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Demonstrating Results Type IV WTG - LLG fault (AB) - BUS 1
11
© 2018 Electric Power Research Institute, Inc. All rights reserved.
CAPE Implementation - Electrocon
• Electrocon has implemented so far the Type IV WTG/
Solar model
• Type III WTG model is under development and will be
presented at the CAPE UGM in June 2018
• Technical paper presented at the CAPE UGM - June 2017
• EPRI and Electrocon have benchmarked the Type IV
WTG model using a 9 bus test system
12
© 2018 Electric Power Research Institute, Inc. All rights reserved.
OneLiner Implementation - ASPEN ASPEN OneLiner v14
13
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Webcast: EPRI Wind/Solar Phasor Domain Short-Circuit Models &
Implementation Status in Commercial Tools
Link https://www.epri.com/#/pages/product/000000003002010940/
14
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Model Validation – 3 Approaches
2. Manufacturer EMT Models
1. Generic EMT Models
15
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Type-III WTG Wind Park Connected to a 230-kV Substation
EMTP Model
Slack: 230kVRMSLL/_0
Vsine_z:VwZ1 V1:1.00/_-0.00
V2:0.00/_102.09
LF V0:0.00/_45.00
LF1 Va:1.00/_0.00
Vb:1.00/_-120.00
VwZ1 Vc:1.00/_120.00
+
6604_LATIGO
230kVRMSLL /_0 + Relay_Transmission
+
Line_LATIGO_3BUTTES
PI
WP_DFIG1
Relay_Wind
+
11847_THREE_BUTTES
DFIG AVM V1:1.00/_0.2
110.022MVA V2:0.00/_-89.8
230kV V0:0.00/_-89.8
Q-control Va:1.00/_0.2
Vb:1.00/_-119.8
Vc:1.00/_120.2
Variable
EMTP-RV Phasor Model
16
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Solar Model Validation with Recorded Data
17
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Validation of Type III Wind Turbine with Manufacturer EMTP Model
18
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Protection Guidelines for Systems with High Levels of
Inverter Based Resources
19
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Protection System Performance Evaluation – EPRI Guidelines Document
• Protection System Performance Evaluation: Study relays response & identify relay misoperation
scenarios on benchmark systems with high renewable penetration.
• EPRI Guidelines Document: Provide recommendations and study practices to protection engineers
when conduction protection studies to prevent relay misoperation/miscoordination
20
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Impact of Inverter-Based Generation on Negative Sequence Based Protection
Typically inverter-based resources are designed Negative Sequence Directional
to suppress negative sequence current partially Reverse fault misinterpreted as forward
or entirely.
Negative-sequence current suppression may
result in the misoperation of legacy negative-
sequence-based protection schemes.
– Negative-sequence overcurrent element
– Negative-sequence-based directional element
Inverter controls define the phase angle
relationship between negative sequence voltage
and current
23
© 2018 Electric Power Research Institute, Inc. All rights reserved.
PSRC WG C24 “Modification of Commercial Fault Calculation Programs for
Wind Turbine Generators”
• Chair: Dr. Sukumar Brahma (NMSU), Vice-Chair: Evangelos Farantatos (EPRI)
• Scope:
• 1) To survey WTG manufacturers to determine what parameters they could provide that could be used by steady state short
circuit program developers in various time frames.
• 2) Use the result of this survey to prepare a report that can be used by steady state program developers to refine their models.
• EPRI has a leading role to the WG. Members include WTG manufacturers (Siemens, GE) and software vendors
(Electrocon, ASPEN, ETAP)
• WG has proposed a voltage controlled current source model with iterative solution
• Input model data:
• Algorithms for generic converter control schemes (EPRI proposal)
• Tabular format (suggested to be provided by manufacturers with non generic converter control scheme)
25
© 2018 Electric Power Research Institute, Inc. All rights reserved.
IEEE PES - NERC Technical Report “Impact of Inverter Based Generation on
Bulk Power System Dynamics and Short-Circuit Performance”
• Joint IEEE PES and NERC TF
• Co-chairs: Kevin Jones & Pouyan Pourbeik
• Target completion: Summer 2018
• Chapter 2: “Large System Impact Issues Related to Large
Penetration of Inverter Based Resources”
• Chapter 3: “Protective Relay Issues Related to Large Penetration
of Inverter Based Resources”
• PSRC CTF34 “Inverter Based Generation Short Circuit Current
Impacts”
• Assignment: “Coordinate/communicate the efforts of the
PES/NERC Low Short Circuit Current Impacts Task Force and
PSRC working groups addressing the issues of inverter-based
resources”
• Chair: Kevin Jones, Vice chair: Gary Kobet
• EPRI is participating in CTF34
26
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Together…Shaping the Future of Electricity
27
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Inverters and Harmonics
17-May, 2018
Sean McGuinness
Principal Technical Leader,
EPRI Intl.
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Introduction
3
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Solar PV Inverters
0.30%
Harmonic Current
0.25%
0.20%
0.15%
0.10%
0.05%
0.00%
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Harmonic Index
4
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Type 3 Wind Turbine - DFIG
Harmonics emit from:
– Grid Side Converter
– Rotor Side Converter
– Induction Generator air-gaps
0.40%
0.25%
0.15%
0.10%
0.00%
5 7 11 13
Harmonic Coherency of Harmonic Voltage Phase Impact of harmonic Harmonic Index
Order Angle distortion
Example Spectrum of 10x2.5 MW Type 4 WTG
>10th Voltage phase angle variation is pretty Destructive interference 0.18%
7
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Impact of Cable Network
Wind and Solar Power Plants tend to have large
collector networks, mostly underground cable
8
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Impact of Cable Network 2
11
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Together…Shaping the Future of Electricity
12
© 2018 Electric Power Research Institute, Inc. All rights reserved.
Modeling for EMT studies
Inverter-Based Resources Webinar Series
Open Distribution
Elecromagnetic Transient Output (Instantaneous Quantities)
1.50
What is EMT?
1.00
0.00
-0.50
-1.00
-1.50
domain
0.60
0.40
2
These are not your Grandfather’s
EMT Studies!
3
Where might you use EMT today?
Classical studies are still done:
Lightning evaluation
TOV/Insulation coordination
Switching/line energization
Transformer energization
Harmonic analysis
SSR (Remember Mojave?)
HVDC/FACTS control design and analysis
Many other fascinating things done with EMT since
1970!
4
Where might you use EMT for IBR
connections today?
Weak Systems (See webinar #3!)
Regions with high chance of power electronic device
interactions (eg. dynamic performance, FRT tests, etc)
SSO phenomena (SSR, SSCI, SSTI, SSTA, SSTMTLA)
Blackstart analysis
Detailed fault current / protection analysis.
Model Validation
Support for Factory Acceptance Testing
Others….
5
Clarifications on Weak Grids
(From the last webinar) “It is very difficult to use SCR to set
planning guidelines and thresholds… a lot of study is
required!”
1. Difficult ≠ Impossible…
2. There are limits for conventionally controlled IBR plants…
Planners need to plan accordingly, use all available screening
and analysis tools, and make judgement calls in absence of
complete information.
3. Mitigation of weak grid issues (incl. using Sync.
Condensers) requires care. You can make new problems for
yourself!
4. Distributed rooftop PV is a whole new world of weak grid
uncertainty… Research is required and is ongoing, but can’t
happen quickly enough!
6
SSCI Phenomena
Currently, any utility connecting IBR near series
compensated circuits is (or should be!!) performing EMT
studies to verify stability.
Examples: PacifiCorp, any ERCOT utility, any Australian utility,
ISONE, SCE, most Canadian provinces, many European
utilities etc…
7
Model Validation
Transient
stability models
may be tested
against highly
detailed (real
code) EMT
models to verify
performance
that can’t be
validated in the
field.
Example from
ISONE:
8
Trend: Large composite EMT models
Increasingly common as tools allow the capability. Current study
examples:
Electranet, South Australia: 30 IBR plants, 700 busses (Entire South
Australian grid) – Sync Condenser sizing study.
South Texas: 22 IBR plants, 7 SVCs, 469 busses – SSCI studies.
ERCOT Panhandle: 25 IBR plants, 435 busses – “weak grid” studies.
(planned) ATC Southwest Wisconsin: 12 Windfarms – dynamic
performance tests
PacifiCorp: 21 IBR models, 50 complex load models – high transfers
with complex RAS, IBR performance critical for new SVC sizing.
HECO: 8 IBR plants and 18 DG PV proxy models – High penetration
dynamic performance evaluation
ISONE: 5 IBR plants and multiple new HVDC ties – High chance of
interaction with very large planned offshore and DC projects.
AEMO (Australian ISO): Long blackstart sequences, very large models.
9
Parallel/Hybrid EMT Simulation
Models can be
parallelized to allow
increasingly large
networks to be
represented:
10
So you need EMT. What should an EMT
generator model look like? Accurate!
Represent full detailed inner control loops of the
converters
Eg. current control, PLL, etc.
Represent all pertinent control features
Eg. FRT, SSCI damping, voltage control, plant control etc.
Represent all pertinent electrical and mechanical
configurations
Eg. transformers, filters, gearboxes, etc.
Represent all pertinent protections
Eg. OV, UV, DC bus, overcurrent, and many others
11
What should an EMT generator model look
like? Usable!
Accessible user control for settings which are adjustable
Reasonable time-step (>10 – 20 us)
Documentation included, and a test case
Self initializing
Multiple instance support (and snapshots?)
Accepts reference values (no hardcoded setpoints!)
Stable interfacing/scaling methods (injection models often not
suitable)
Others… see:
http://www.electranix.com/publication/technical-memo-pscad-model-requirements/
12
Electranix Corporation
But Andrew, that’s old material!
Yes, and manufacturers have (mostly) listened!
Current model quality is getting better
New IEEE working group (“Use of Real-Code in
EMT Models for Power System Analysis”) on “real
code” model development. All plusses!!
State of the art “real code” models are now
becoming widespread.
You still need to check model quality.
BUT… Some manufacturers are now complaining
that their beautiful generator models are being
used with overly simplistic system models.
13
Doing an EMT study?
What can you tell from this
model?
Small Signal control stability
(SCR Tests) (yes, sometimes) P = 99.43
Vsys_LGpk
Q = 32.75
SSCI? (no) V = 1.051
+ +
V
R=0
FRT? (no)
Beautiful A
Detailed Model
V Rsource Lsource
Fault
necessarily)
Insulation coordination? (no)
Harmonics? (no)
Lightning/TOV/TRV/etc? (no)
14
Dynamic performance models
If you want to correctly evaluate these things, you need:
Sufficiently big system model (How big? Good question, but maybe
bigger than you think!)
Frequency dependent representation of transmission lines
Series Capacitors (and sometimes series capacitor arresters and
protection)
Transformer saturation, tap settings, winding configurations, etc.
Detail for nearby power electronics (HVDC, wind, SVC, etc.)
Suitable load representation
Synchronous machine dynamic representation
Shunt capacitors and reactors, neutral reactors, etc.
Suitable system equivalent (multi-port or dynamic if necessary)
Depending on the study, you may also need:
Station and line surge arresters
Protection, SPAR/TPAR schemes, RAS, SPS data, other stuff…
15
Dynamic performance models
The specific system model you need depends on what
problems you’re looking for…
Tools are available to simplify model development and extract
available data from the powerflow and transient stability tools
(full disclosure… Electranix sells some of these – E-Tran suite of tools, some of which
are licensed as PSS/E modules)
Initialization, data and library management is required…
Even with these tools, there is a lot of extra data to obtain and
enter. There is a lot of work to do!
16
Example
System
Model: P = 99.43
Vsys_LGpk
Q = 32.75
V = 1.051 V
+ + R=0
Beautiful A
Detailed Model
V Rsource Lsource
Fault
17
(example system equivalent)
18
What is the future for EMT?
Interconnected/harmonized system models
(cross-platform, cross-genre models)
Real code models
Very large parallelized EMT models for
dynamic performance evaluation
Regional library development for EMT
system models
As IBR and DG penetration continues to
rise and fossil fuel based generation
declines, need for new control concepts
(and detailed models) will rise too.
Increasing adoption of EMT as part of
routine planning
NERC-NATF-EPRI-UVIG
PUBLIC 2
Series Capacitors in ERCOT
Cross (x2)
Gauss (x2)
Kirchhoff (x2)
PUBLIC 3
South Texas SSR Event (2009)
PUBLIC
Induction Gen Effect and Control Interaction
PUBLIC 5
Types of SSR
• Transmission Project-Related
Studies
• Annual Assessment
PUBLIC 7
ERCOT Study Process
(Same process for all generators)
TOPOLOGY CHECK
0.25
Impedance (ohms)
0.20
0.15
Resistance
0.10
Reactance
0.05
‐0.05
5 9 13 17 21 25 29 33 37 41 45 49 53 57
Frequency (Hz)
STUDY P1 : Graphs
PWF (MW) QWF (MVAr)
250
200
150
100
50
-50
-100
x 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
PUBLIC 8
Topology Test: Counting Contingencies
345 kV B
Outages “A”
and “B” make
138 kV
the generator
A “radial” to the
series capacitor
GenY ~
PUBLIC
Study Process
TOPOLOGY TEST
0.30
0.25
Impedance (ohms)
0.20
FREQUENCY SCAN
0.15
Resistance
0.10
Reactance
0.05
0.00
5 9 13 17 21 25 29 33 37 41 45 49 53 57
‐0.05
Frequency (Hz)
TIME DOMAIN STUDY
P1 : Graphs
PWF (MW) QWF (MVAr)
250
200
150
100
50
PUBLIC 10
How Do We Run Frequency Scans?
Measure Thevenin impedance at
different frequencies 5 ~ 55 Hz
Network
System-Side Generator-Side
Frequency Scan Frequency Scan
Don’t forget to
simulate
contingencies!
11
PUBLIC
How Do We Run Frequency Scans?
+
Total System+Turbine Scan Total System+Turbine Scan
140 60
120
R X 40 R X
100
80
=
20
60
40
0
20 25 27 29 31 33 35
0
‐20
0 10 20 30 40 50 60
‐20
‐40 ‐40
‐60
‐60
12
Here, resistance is positive where reactance crosses from negative to
positive, so anticipate no SSR concern
PUBLIC
Frequency Scans have limitations
(Small Signal vs Large Signal)
• Observed during SSR study:
– Frequency scan indicated positive resistance GOOD
– Non-fault EMT indicated stable response GOOD
– Fault EMT indicated SSR and tripping BAD!
• What happened?
– Frequency scans are linearized around an operating point. Thus
good for small signal but not for large signal.
– IBR are much more non-linear than synchronous generation
• Different operating modes and switches in controls
PUBLIC
What about sensitivities? (Some observations…)
• Large impact:
– Transmission Outages
– Nearby generators on/offline
• Moderate impact:
– Number of turbines online and real / reactive dispatch
– Fault / faultless outages
– Series capacitor bypass configurations
• Lesser impact:
– Switched reactive shunts
PUBLIC 15
Industry Progress
PUBLIC 16
Protection vs Mitigation – ERCOT’s Practice
PUBLIC 17
Issues Locating Very Close to Capacitors
• Recently,
Renewable Resources are connecting
closer to series capacitors, even tapping
series-compensated lines themselves
PUBLIC
“Energy Accumulation” and VRT Tripping Issues
PUBLIC 19
Questions?
PUBLIC 20