Transcript EED`s supporting capabilities and investments

Non-iterative voltage stability
analysis methods and prototype
software for multi-path rating
Yuri V. Makarov
• WECC JSIS Meeting
• Salt Lake City, UT
• September 10, 2014
Project Team
Dr. Bharat Vyakaranam – Research Engineer, Power
Systems, PNNL
Dr. Da Meng – Research Engineer, Power Systems,
PNNL
Dr. Pavel Etingov – Research Engineer, Power Systems,
PNNL
Dr. Tony Nguyen – Research Engineer, Power Systems,
PNNL
Dr. Di Wu - Research Engineer, Power Systems, PNNL
Dr. Zhangshuan (Jason) Hou – Exploratory data
analyses and uncertainty quantification, PNNL
Dr. Shaobu Wang - Research Engineer, Power Systems,
PNNL
Dr. Steve Elbert – High Performance Computing, PNNL
Dr. Laurie Miller – Research Engineer, Power Systems,
PNNL
Dr. Yuri Makarov – PM, Chief Scientist, Power Systems,
PNNL
Advisors:
Dr. Zhenyu (Henry) Huang
Dr. Ruisheng Diao
Dr. Mark Morgan
Acknowledgements: DOE ARPA-E (Tim Heidel and Sameh
Elsharkawy) and DOE OE Office (Gil Bindewald)
Overview - 1
Research Objectives
New non-iterative methods for
multi-parameter voltage
stability assessment (VSA) in
near-real-time.
Multi-path rating application.
Answers will be given:
 How far the system is from
instability and blackout?
 What are the most critical
contingencies and system
elements?
 What needs to be done to
increase the security margin in
real time?
 What is the time remaining for a
possible violation? - Future
Voltage stability boundary of a simple system
and its projections. Source: Hiskens and Davy
April 1, 2016
3
Overview - 2
Background/Problem:
Different parts of the VS boundary (VSB) correspond to increasingly variable
stress directions caused by changing load-generation patterns, contingencies,
market forces, cooperation between system operators, variable generation, etc.
Computational time becomes critically
important for:
Path 1
Real-time analyses
Massive contingency screenings
Simulations of blackouts and cascading
Probabilistic methods
Synchrophasor-based applications, and
Traditional methods (e.g., continuation
power flow - CPF) are:
Path 3
Path 2
April 1, 2016 power flow process: π – predictor; σ – corrector.
Continuation
Computationally intensive,
Limited by a few stress directions
Based on simplifications,
Sensitive to initial guesses.
Overview - 3
Benefits and Impacts:
Enhanced situation awareness
Early detection of system instability,
Improved reliability
Actionable information,
Prevention of system blackouts, and
Better utilization of transmission assets.
d1
Hd
ξd
D0
Other benefits:
d2
Security Region
d3
April 1, 2016
VSB visibility for multiple paths and
contingencies
Developing real-time and HPC
applications
Accurate and flexible quantification of the
VS margins
Wide-area view on voltage stability
Potential for predictive/preventive control
Potential for close-loop automatic
emergency control systems.
5
Security Margin and Control Direction
d1
Hd
ξd
D0
d2
Security margin ║ξd║
provides situation
awareness
Control vector ξd
provides actionable
information
Constraints applied to
control parameters and
their priorities can be
incorporated.
Security Region
d3
April 1, 2016
6
Approach -1
We are using powerful methods
to explore voltage stability
boundary (VSB)
1
Orbiting method
Each iteration produces a new
VSB point
We do not have to repeat
continuation power flow for
each VSB point!
Is very fast and accurate
2
3
Path 2
CPF
2-D “Slice” of n-D Voltage Stability Boundary
Path 1
Providing Connectivity With PowerWorld
Input: Three System Models Tested
Central America
Interconnection of Panama, Costa Rica, Honduras, Nicaragua, El Salvador,
and Guatemala systems
1985 buses
2298 branches
California ISO
3535 buses
4402 branches
Western Electricity Coordinating Council planning model
19331 buses
22946 branches
8
Simulation Results- Central America
Voltage Security Region in Injection Space - Load at two buses
1.2
in 100 MWs
1
0.8
Load at Bus SID-22
0.6
0.4
0.2
0
-0.2
 CPF run for one VSB point
• 7.6963 s
 BOM run
• 0.1655 s
Simulation Times
Systems
Continuation
power flow
Boundary
orbiting method
Average Time Per Run
(s)
Average Time Per Run
(s)
WECC2014
(19331 buses)
191
16.64
CAISO
(3535 buses)
9.6306
1.1010
Central America
(1985 buses)
7.6963
0.1655
Accuracy Comparison With PowerWorld
Stress
direction
11
Non-iterative Method
PowerWorld
Accuracy
Sink Load
(MW)
Sum(Sink)
(MW)
Sink
Load(MW)
Sum(Sink)
(MW)
%
1
740.08
740.08
738.11
738.11
0.28
2
1188.9
1188.9
1192.8
1192.8
0.82
3
260.58
260.58
261.61
261.61
0.4
4
744.56
1002.4
746.95
1007.0
0.46
Connections with Previous, Existing, and Future
Funded Projects and Outreach Activities
University of Sydney,
Australia, ARC grant
X-ray theorem and
Delta-plane method,
1993-1997
PNNL
DOE ARPA-E project
Non-wire methods
FY 2013-2015
PNNL
LDRD project
Further development
of Non-iterative
voltage stability
analysis method
Cost Sharing
PNNL
CEC/ CERTS /EPG
project
Voltage stability
orbiting procedure
PNNL
DOE OE project
Wide-area security
region
PNNL
BPA project
Wide-area security
nomogram
PNNL
DOE OE project
Non-iterative
voltage stability
Further outreach, technology
transfer & commercialization:
Utilities and ISOs: BPA, …
Software Vendors: PowerWorld,
…
Consulting Companies: Quanta
Technologies, …
ARPA-E 0670-4106
Multi-path Near-Real-Time Path Rating:
General Project Progress and Updates
Team:
PNNL (Prime): Henry Huang, Ruisheng Diao, Shuangshuang Jin, Yuri
Makarov, Yousu Chen
Quanta Technology (Sub-Prime): Guorui Zhang
PowerWorld: James Weber
Bonneville Power Administration: James Wong, Brian Tuck
13
Transmission congestion cost
Incur significant economic cost
2010: >$1.1 billion congestion cost at New York ISO [1]
2010: $ 1.43 billion congestion cost PJM-wide [2]
14
[1] NYISO, “2011 Congestion Assessment and Resource Integration Study”, March 2012
[2] PJM, “Congestion and the PJM Regional Transmission Expansion Plan”, Dec. 2011
Means of congestion management
Three traditional means of congestion management (all
require capital investment) [3]:
Build more generation close to load centers.
Reduce load through energy efficiency and demand reduction programs.
Build more transmission capacity in appropriate locations.
Near-real-time approaches:
Generation redispatch (additional cost)
Dynamic Line Rating (DLR), thermal limited
Validated at RTE, France and Oncor, TX
Real-time path rating, security/stability limited
Validated concept at BPA, CAISO and ERCOT
No tools available due to intensive computational requirements using existing
techniques
15
[3] 2012 National Electric Transmission Congestion Study. David Meyer, U.S.
Department
of Energy, August 2012.
Real-time path rating
Current Path Rating Practice and Limitations
Offline studies – months or a year ahead of the operating season
Worst-case scenario
Ratings are static for the operating season
 The result: conservative (most of the time) path rating, leading to
artificial transmission congestion
Real-Time Path Rating
On-line studies
Current operating scenarios
Ratings are dynamic based on real-time operating conditions
 The result: realistic path rating, leading to maximum use of
transmission assets and relieving transmission congestion
16
Benefits of real-time path rating
Increase transfer capability of existing power network
and enable additional energy transactions
Reduce total generation/consumer cost
Avoid unnecessary flow curtailment for emergency
support, e.g. wind uncertainties
Enable dynamic transfer
Enhance system situational awareness
17
Technical Approach and Objective
Technology Summary
1. Develop HPC based transient and voltage
stability simulation with innovative
mathematical methods
2. Develop HPC based real-time path rating
capability with predictability and
uncertainty quantification
3. Develop advanced congestion
management methods with hierarchical
coordination and optimized control
4. Demonstrate the non-wire method on a
commercial software platform with reallife power system scenarios
Proposed Targets
Metric
State of the
Art
Proposed
Simulation speed
3-5 times
slower than
real time
10-20 times
faster than real
time
Path rating study
internal
Months
<10 minutes
Uncertainty
quantification
No
Yes
Asset utilization
Conservative
Enhanced
Project management and coordination
19
Industry Advisory Board:
José Conto - Principal Engineer, Electric Reliability Council of Texas
Anthony Johnson - Consulting Engineer, Southern California Edison
Xiaochuan Luo - Technical Manager, ISO-New England
Joshua Shultz - TVA
Dede Subakti - Director, Operations Engineering Services, CAISO
Current activities
Project team actively working on recent deliverables
Fast dynamic simulation
Implemented full Y matrix for network solution
Tested and compared different linear solvers
Non-iterative voltage stability method
Improved MATLAB code for better accuracy and speed
Accuracy validated against a commercial package, PowerWorld
Simulator
Developed multiple path rating studies procedure
Defined interface functions for software integration
20
Chart 1
Chart 2
Final products
d1
North of John Day vs. COI + NW/Sierra or PDCI Flow
(Summer 2008 N-S Nomogram)
Hd
4900
4800
Midpoint - Summer Lake
4700
400 MW East to West
MW East to West
4600
ξd
D0
4500
4400
4300
4200
4100
Midpoint - Summer Lake
400 MW West to East
PDCI or COI + NW/Sierra Flow (MW)
4000
3900
Midpoint - Summer Lake
0 MW
3800
d2
3700
3600
3500
Midpoint to Summer Lake flows are
shown in increments of 200 MW
3400
Security Region
3300
d3
3200
3100
3000
2900
2800
2700
2600
2500
Solid lines are for COI + NW/Sierra limits and
Dashed line is for PDCI limit.
2400
7000
7100
7200
7300
7400
7500
7600
North of John Day Cutplane Flow (MW)
7700
7800
7900
8000
n11d1  ...  n1n d n  Ld ,1

n21d1  ...  n2n d n  Ld ,2

...
n d  ...  n d  L
mn n
d ,m
 m1 1
Questions?
24