Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
HELICS for Integrated Transmission, Distribution, Communication, & Control (TDC+C) Modeling SETO Workshop on Challenges for Distribution May 17, 2019 Washington, DC
Presented by: Bryan Palmintier
Henry Huang, Liang Min, Jason Fuller, Philip Top, Dheepak Krishnamurthy, Shri Abhyankar, Manish Mohanpurkar, Kalyan Perumalla, David Schoenwald
NREL/PR-5D00-73977
Grid modernization requires integrating multiple
infrastructures…
Transmission Distribution
Control & Communication
Transportation,Water,
Gas, etc.
And we have many, well trusted tools to model each…
PSLF PSS®E
Transmission
WindMil CYME
Synergi
Distribution
Control & Communication
SCEPTRE MATLAB
Transportation,Water,
Gas, etc.
SAInt
However they are largely used within their own silos of excellence.
Transmission
PSLF PSS®E
Control & Communication
SCEPTRE MATLAB
Distribution
Transportation,Water,
Gas, etc.
WindMil CYME
Synergi
SAInt
HELICS enables easily bringing together two or more existing tools, exchanging data as time advances, to form a tightly integrated co-simulation.
Transportation,Water,
Gas, etc.
SAInt
PSLF PSS®E
Transmission
WindMil CYME
Synergi
Distribution
Control & Communication
SCEPTRE MATLAB
MATPOWER Transmission/Bulk:
AC Powerflow, Volt/VAr
FESTIV Runtime plug-in
Bus Aggregator Bus Aggregator
Bus Aggregator
bus.py bus.py ...
ZMQ
MPI
IGMS-Interconnect
bus.py bus.py
Hom
e &
Appli
ance
Phy
sics
Hom
e &
Appli
ance
Phy
sics
Hom
e &
Appli
ance
Phy
sics
Alte
rnat
e Di
stribu
tion
Mod
elTim
eser
ies, e
tc. HT
TP
HTTP
HTTP
Scen
ario
Aut
omat
ion
Tran
smiss
ion
Distrib
ution
HELICS™: Hierarchical Engine for Large-scale Infrastructure Co-Simulation Project funding: GMLC 1.4.15
Capabilities: • Scalable: 2-100,000+ Federates • Cross-platform: HPC (Linux), Cloud,
Workstations, Laptops (Windows/OSX) • Modular: mix and match tools • Minimally invasive: easy to use
lab/commercial/open tools • Open Source: BSD-style. • Many Simulation Types:
• Discrete Event • QSTS • Dynamics
• Co-iteration enabled: “tight coupling” • APIs: C++,. C, Python, Java, Matlab,
Julia, FMI
Best of Existing Cosim
FESTIV: ISO Markets, UC & AGC
GRID
Lab-
DDi
stribu
tion
Powe
rflow
,
GRID
Lab-
DDi
stribu
tion
Powe
rflow
,
GRID
Lab-
DDi
stribu
tion
Powe
rflow
,
...
ISO
Build
ingAp
plian
ce
FNCS/ GridLAB-D
FSKit/ GridDyn
IGMS/ FESTIV
Use Case Requirements
New Platform Design
End-UseCont
rol
Markets
Communication
DistributionTransmission
“TDC Tool”
B. Palmintier, et al., “Design of the HELICS High-Performance Transmission-Distribution-Communication-Market Co-Simulation Framework,” Workshop on Modeling and Simulation of Cyber-Physical Energy Systems, Pittsburgh, PA, 2017.
v2.0.0 available now at https://www.github.com/
GMLC-TDC/HELICS-src Co-simulation Engine
Grid
DDER
Trans port ICT Water
Scalable, High-performance co-simulation to combine best-in-class tools for breakthrough grid modernization simulation and analysis
6
Computational Integration Workflows
Slide adapted from Dr. Wes Jones, NREL
Co-simulation “consists of the theory and techniques to enable global simulation of a coupled system via the composition of simulators. Each simulator is a black box mock-up of a constituent system, developed and provided by the team that is responsible for that system.”[1]
[1] Gomes, Cláudio et al. “Co-simulation: State of the art.” CoRR abs/1702.00686 (2017): n. pag. Figures: Gomes, Cláudio et al. “Co-Simulation: A Survey.” ACM Comput. Surv. 51 (2018): 49:1-49:33.
Co-modelling is “w[h]ere models are described in a unified language, and then simulated.”[1]
Computational Integration Workflows
[1] Gomes, Cláudio et al. “Co-simulation: State of the art.” CoRR abs/1702.00686 (2017): n. pag. Figures: Gomes, Cláudio et al. “Co-Simulation: A Survey.” ACM Comput. Surv. 51 (2018): 49:1-49:33.
Slide adapted from Dr. Wes Jones, NREL
“Cyber”-Physical Simulation e.g. Transmission-Distribution-Market
▶ Physical Data (Values) v Voltage, Frequency, Current
▶ Market Data (Messages) v Measured Load, LMPs
M Market Dispatch
Actual Generation
RT Prices (LMP)
Measured Load
T Transmission
DGridLAB-D
Load per phase
Voltage
DGridLAB-D
DGridLAB-D
DGridLAB-D
D Distribution
Large-scale DER-Market Interactions
NREL’s Integrated Grid Modeling System
GR
IDLa
b-D
Dis
tribu
tion
Pow
erflo
w,
Hom
e &
Appl
ianc
e Ph
ysic
s
Alte
rnat
eD
istri
butio
n M
odel
Tim
eser
ies,
etc
. HTT
P
(IGMS) provides a full-scale co-simulation with transmission-level markets, 1000s of distribution feeders, and 1Ms of DERs
FESTIV: ISO Markets, UC & AGC
MATPOWER Transmission/Bulk:
AC Powerflow, Volt/VAr
FESTIV Runtime plug-in
Bus AggregatorBus Aggregator
Bus Aggregator
bus.py bus.py ...
ZMQ
MPI
IGMS-Interconnect
bus.pybus.py
GR
IDLa
b-D
Dis
tribu
tion
Pow
erflo
w,
Hom
e &
Appl
ianc
e Ph
ysic
s
GR
IDLa
b-D
Dis
tribu
tion
Pow
erflo
w,
Hom
e &
Appl
ianc
e Ph
ysic
s
...
HTT
P
HTT
P
Scen
ario
Aut
omat
ion
ISO
Tran
smission
Distribu
tion
Building
App
lianc
eNew for IGMS v3: HELICS
co-simulation
9
Adding Controllers… e.g. Control Architecture Scaling & Performance
▶ Physical Data (Values) v Voltage, Frequency, Current
▶ Market Data (Messages) v Measured Load, LMPs
▶ Controller Data (Messages) v Sensor Readings, Control Signals
M Market Dispatch
Actual Generation
RT Prices (LMP)
Measured Load
T Transmission
DGridLAB-D
Load per phase
Voltage
DGridLAB-D
DGridLAB-D
DGridLAB-D
D Distribution
C Controller
Sensor Data
Sensor Data
Control Signals
Control Signals
Novel T&D Control Architecture Design: Predictive State Estimation & Machine Learning Control Grid Sim: Entire Island of Oahu, HI with >1M electric nodes.
10
Keeping the wires uncrossed
Inv 1&2 Substation
Legend Voltage Regulator
Capacitor Bank
PV Inverter RecloserController
Actual Deployment
Inv 1&2 Substation
Legend Voltage Regulator
Capacitor Bank
PV Inverter Inv 1&2
Substation Recloser
Legend Voltage Regulator
Capacitor Bank
PV Inverter RecloserController
Software Simulation
Keeping the wires uncrossed
Actual Deployment
Inv 1&2 Substation
Legend Voltage Regulator
Capacitor Bank
PV Inverter Recloser
Software Simulation
v Delays v Random drops v Other message effects (e.g.
packetization) v And more
▶ No changes to domain models
M Market Dispatch
Actual Generation
RT Prices (LMP)
Measured Load
T Transmission
DGridLAB-D
Load per phase
Voltage
DGridLAB-D
DGridLAB-D
DGridLAB-D
D Distribution
C Controller
Sensor Data
Sensor Data
Control Signals
Control Signals
CommFilter
CommFilter
Comm Filter
Comm Filter
… and Simple Communication e.g. Design-stage Cybersecurity Evaluation
▶ Built in “Filters” for
1313
Novel T&D Control Architecture Design: Predictive State Estimation & Machine Learning Control Grid Sim: Entire Island of Oahu, HI with >1M electric nodes.
1. Control signal spoofing
2. Control node compromise 3. Sensor data spoofing
4. Communication Denial of Service
X
5. Communication Latency Margin
Or Detailed Communication e.g. Protocol Comparison for Situational Awareness
▶ Full communication simulation: v Shared bandwidth
v Network Specific Vulnerabilities
v Potential Tools: ns-3, Opnet++, SCEPTRE, etc.
▶ No changes to domain models
M Market Dispatch
Actual Generation
RT Prices (LMP)
Measured Load
T Transmission
DGridLAB-D
Load per phase
Voltage
DGridLAB-D
DGridLAB-D
DGridLAB-D
D Distribution
1414
C Controller
Sensor Data
Sensors
C Communication
reroute
reroute
Controls
Projec
Protocol/Full-Stack Performance
t Ex: SuNLaMP Hybrid Comms
Ex: Power-Comm. Emulation
Some Other Use Cases
60.05 0% PV 10% PV 20% PV Large-scale DER-Market Sim 30% PV
60 • 35k feeders
frequ
ency
(Hz) 59.95 • WECC-240 trans.
59.9 • 25M homes 59.85 • Simplified CAISO-
59.8 0 1 2 3 4 5 6 7 8 9 10 style Market
Time (sec)
T&D frequency stability with high DER
Inv 1&2 Substation
Legend Voltage Regulator
Capacitor Bank
PV Inverter Recloser
Inv 1&2 Substation
Legend Voltage Regulator
Capacitor Bank
PV Inverter Recloser
Noon 3pm 6pm 9pmADMS Testbed and other PHIL Figure from Trevor Hardy, PNNL
15
HELICS Interfaces to Domain Simulators
Not exhaustive lists.
16
▶ Growing mix of tools
▶ Enable large-scale interdependency all-hazards studies: scale to 100,000+ domain simulators
▶ Diverse simulation types: v Continuous, discrete event,
time series
v Steady-state/dynamic/transient
v Any energy system
▶ Support standards: HLA, FMI, …
▶ APIs: C++,. C, Python, Java, Matlab, Julia, FMI
PSLF Windmil
Gas-Pipeline
Energy+ (buildings)
Transportation
T D
C More
GridDyn InterPSS
MATLAB (PST, MATPOWER)
FESTIV
GridLAB-D
MATLAB
NS3
HELICS built-in Comms Sim
OpenDSS
Existing Ongoing Waiting
Powerworld CYME
SCEPTRE
Thank You
Bryan Palmintier: [email protected]
HELICS v2.0.0 available at https://www.github.com/GMLC-
TDC/HELICS-src
NREL/PR-5D00-73977
Henry Huang, Liang Min, Jason Fuller, Philip Top, Dheepak Krishnamurthy, Shri Abhyankar, Manish Mohanpurkar, Kalyan Perumalla, David Schoenwald
This work was authored in part by Alliance for Sustainable Energy, LLC, the manager and operator of the National Renewable Energy Laboratory for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by U.S. Department of Energy Offices of Electricity and of Energy Efficiency and Renewable Energy Solar Technologies Office through the Grid Modernization Initiative. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, 17 worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.