arpa-e micro-synchrophasor project: modeling and...
TRANSCRIPT
ARPA-E micro-Synchrophasor Project: Modeling and Validation
Emma Stewart, Chuck McParland & Sila Kiliccote
Grid Integration Group, LBNL February 21th 2014
2
Acknowledgment: The information, data, or work presented herein was funded in part by the Advanced Research Projects Agency-Energy (ARPA-E), U.S. Department of Energy, under Award Number DE-AR0000340. Disclaimer: The information, data, or work presented herein was funded in part by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
3
SCE: SONGS
• Imbalance of supply and demand anticipated in southern California due to SONGS shutdown.
• Focus on South of Lugo area served by SCE
• Issues: SONGS off is N-1 – System needs rebalanced
such that SONGS off is now N and the area can cope with N-1, N-2 and N-1-1
– Currently SONGS + high load + outages could result in N-3 (very bad)
4
PG&E Operations in Concord, California
Why is distribution “invisible”? • It worked • No major incidents to spur
change – most major outages have been caused by transmission
• What could happen? – Two way power flow – Its becoming an active
resource – No real insight into the impact
of changing power flows on transmission and distribution
5
Review of Distribution Simulation Tools • Review of distribution tools in general
– Steady state, dynamic, transient packages • Review of what are positive features for future software
development in distribution planning • Comparison of available tools for application development such as
state estimation and dynamic device characterization • Important other features
– Interfaces with real time data – Data input/output capabilities – GUI’s
!
6
Planning Tool Selection to Address Utility Concerns Issue Timeframe So.ware op1ons Opera1ons Op1ons FIDVR Milliseconds, sub-‐cycle None Data input to
opera:ons
Dynamic Load Behavior Similar to FIDVR CymDist, PSCAD, DEW, DigSilent
Data input to opera:ons
Dynamic Inverter Behavior
Milliseconds, Seconds CymDist, PSCAD, DEW, DigSilent
Network model representa:on
Controllable inverter behavior (Volt/Var)
Seconds, Minutes, Hours
CymDist, DEW, DigSilent
Data input to opera:ons
Transient switching impacts (microgrids
Subcycle PSCAD, PSS/Sincal, DigSilent
Not opera:on issue – preventa:ve in planning and protec:on analysis
Load flow – reverse power, voltage profiles etc
Hours, minutes, steady state
CymDist, DEW, SynerGEE Electric
Data input to opera:ons
Protec:on in high penetra:on scenarios
Milliseconds DEW, CymDist, SyenrGEE Electric
Not opera:onal issue
7
Preliminary Modeling Tools Review Conclusions • Preliminary observations from analysis of T&D modeling tools:
– The existing models are not designed to simulate the operation of large numbers of distributed resources.
– Model data standardization and validation is essential – Highly variable PV systems must be analyzed in a detailed
dynamic and steady state simulation. – As more detailed inverter control schemes become widely
available to customers, the impact must be justifiable and proven through simulation.
– Availability of measured data is essential for validation and lacking in many cases.
– No single software package can achieve all analysis objectives. • Dynamic and steady state packages are advantageous together • Transient can be separate
8
µPMU Pilot Site Development at LBNL
• Pilot site at LBNL for µPMU devices
• Installing µPMU devices in 4 locations at LBNL – from substation to Building 90
• Will measure high fidelity phase angle data, voltage and current to validate the LBNL CymDist Model
• Challenges: communications, calibration, and commissioning
9
LBNL Model Development & Analysis Plan • Translated schematic from SKM and one-lines to CymDist
– Converges with estimated load • Enhancing substation model for feeder to bldg 90 • Aggregate isolated unmeasured portions of lab distribution system • Analysis plan
– Validation and verification process against standard and uPMU angle data – Baseline studies incluing load flow, fault current and long time step dynamics – Advanced studies of potential application inc high renewable penetration,
transient impacts
BLDG 90
SW STA A6 FEED FROM GP
10
Model Validation with Improved Data
• We often model and assume distribution lines flow (top graph) and voltage is represented by the blue line
• In reality it looks like the red line
• Distribution model may have been constructed assuming traditional flow (all one direction)
• Present and future needs more fidelity at line and phase level
11
Distribution versus Transmission Validation • Need for a proactive approach to distribution model validation
– Transmission task force was in response to incidents – Need to build a distribution system model validation process from
lessons learned in transmission • Challenges:
– Unbalanced non-symmetrical systems, harder to solve – transmission assumption (R/X ratio not true) – More components – Dispersed load, many dynamic types – Less measured data, time synchronization, more time series
changes • Operations models in early stages of implementation
– Create a process to validate planning and operations models concurrently
12
Data Standardization • CIM: Common Information
Model – Defines the structure of data in a
logical model • Originally for the exchange of
data between EMS and GIS – Enabling the integration of
network models at both transmission and distribution levels
– Expanded to include distribution extensions
• Does not include all data required for increasingly complex analyses on distribution systems, but can be extended for use cases and further development
13
Model Validation Process
14
Analysis Plan for 20+ Feeders
• Get models (SynerGEE, utility format) • Determine issues of concern with utilities, obtain any
existing measured data – select tools to use for each utility • Transfer model data to selected tools • Validate models in measured condition • Determine baseline conditions to perform for all models for
comparison • Perform baseline analysis – load flow in varied load/gen
conditions • Run key diagnostic issues on validated planned models • Report on usefulness of data in planning models and
validation
15
Summary • Challenges In Modeling and Analysis
• Data format from utilities • Model, measured, other • Transmission representation
• Current issues in the distribution system • Volume of unknowns • Lack of control and visibility • Inaccurate and un-validated modeling
• There are big challenges in distribution systems simulation and analysis
• Observability – if we cannot see a resource how can we get the most benefit from it?
• Appropriate accurate analysis – validated models and using the correct tools and applications