Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
( ) Z , V , X I YV =
1529pk - 128
Copyright P. Kundur
Mathematical Formulation
At each equilibrium point, Z=Z
i
and the system operating condition
is obtained by solving the following set of nonlinear algebraic
equations:
As time progresses, the slow control devices operate and the
values of Z change. The above set of nonlinear algebraic equations
is solved every time the values of Z change.
( )
( )
i
i
Z V, X, I YV
Z V, X, f 0
=
=
1529pk - 129
Copyright P. Kundur
1529pk - 130
Copyright P. Kundur
Powerflow
Solver
VSAT Structure
1529pk - 131
Transient Stability Assessment Tool (TSAT)
Copyright P. Kundur
1529pk - 132
Copyright P. Kundur
Transient Stability Assessment (TSA)
Time-domain simulations essential
modeling detail and accuracy
Sole dependence on time-domain simulations has severe limitations
high computational burden
no stability margin/sensitivity information
requires considerable human interaction
Supplementary techniques for speeding up and automating overall
process
Methods available for deriving useful indices
Transient Energy Function (TEF)
Signal Energy Analysis
Extended Equal Area Criterion (EEAC)
1529pk - 133
Copyright P. Kundur
A Practical Tool for TSA
Overall architecture similar to that of VSA
Time-domain program, with detailed models and efficient solution
techniques, forms simulation engine
EEAC used for screening contingencies, computing stability margin,
stability limit search, and early termination of simulation
Prony analysis for calculation of damping of critical modes of
oscillation
A powerflow dispatcher and solver for finding the stability limit
a fully automated process
No modeling compromises;
can handle multi-swing instability
1529pk - 134
Copyright P. Kundur
EEAC
Integrates the dynamic response in the multimachine space, and
maps the resultant trajectory into a set of one-machine-infinite-bus
planes
By applying complementary cluster center of inertia (CCCI)
transformations
Keeps all dynamic information in the multimachine space
Stability analysis can be quantitatively performed for the image OMIB
systems
Has the same accuracy and modeling flexibility
Fast, quantitative
1529pk - 135
Copyright P. Kundur
EEAC
Loss of transient stability in a power system always starts in a
binary splitting of generators:
Critical cluster of generators
Rest of the system
At any given point in the
time-domain trajectory of
the system, the system
can be visualized as a
one-machine-infinite-bus
(OMIB) system
1529pk - 136
Copyright P. Kundur
EEAC
The classical equal area criterion can be extended to the visual OMIB
system
Stability margin of the system is defined as
( )
( )
>
<
= q
d a
a
a d
a d
d
a d
A A
A
A A
x
A A
A
A A
x
unstable is system the if 100
stable is system the if 100
Thus, -100 s q s 100, and
q> 0 if the system is stable
q s 0 if the system is unstable
q can be used as a stability index
1529pk - 137
Copyright P. Kundur
Use of EEAC Theory
Contingency screening
stability margin gives an indication of the relative severity
Corrective measures for maintaining secure system operation
critical cluster of generators (CCG) provides valuable information
Power transfer limit search
stability limit can be determined in four iterations using stability margin
each iteration involves a detailed simulation and computation of stability
index
1529pk - 138
Copyright P. Kundur
Results - Test System
System description
BC Hydro system
1430 buses
186 generators
4 HVDC links
Interface
GMS and PCN output
Base case transfer = 3158 MW
Contingency
Three phase fault at GMS 500 kV bus
Tripping of one of two 500 kV lines
from GMS to WSN
1529pk - 139
Copyright P. Kundur
Limit Search Results
1529pk - 140
Copyright P. Kundur
Speed Enhancement: Parallel Processing
Code parallelization
differential equations easily parallelized, but not network equations
speed-ups limited by serial slowdown effect
up to 7 times speed-up can be achieved with 20-30 processors
not an effective way
Conventional serial computers offer much faster computational
per-CPU
For multiple contingencies
perform initialization only once
run contingencies on multiple processors - one processor per
contingency
1529pk - 141
Copyright P. Kundur
TSAT Structure
Powerflow
Dispatcher
Time-Domain
Simulation
Stability
Indices
Increase
Transfer
Remedial
Measures
Must Run
Contingencies
Transaction
Definitions
Security Limit?
Sufficient
Margin?
STOP
Yes
Yes
No
Full
Contingency List
Solved Powerflow
+
Dynamic Data
Contingency Screening &
Ranking (EEAC)
No
1529pk - 142
Copyright P. Kundur
Computational Performance of DSA
Target cycle time from capture of state estimation to completion
of security assessment for all specified transactions:
20 minutes
TSA and VSA functions performed in parallel
distributed processing on separate CPUs
This can be readily achieved with low cost PCs
1529pk - 143
Copyright P. Kundur
Computational Speed of DSA (cont'd)
Voltage Stability Assessment:
- screening 300 contingencies 20.0 secs
- detailed security analysis 1.2 secs
with 20 critical contingencies
- one transfer limit search 12.0 secs
Transient Stability Assessment:
- screening 100 contingencies 75.0 secs
- 10 second simulations with 75.0 secs
10 critical contingencies
- one transfer limit search 120.0 secs
- total time for complete assessment < 5 mins
Power System model with 4655 buses, 156 generators, using 1.7 GHz,
Pentium 4 PC with 256 MB memory
1529pk - 144
Copyright P. Kundur
Future Trends in DSA: Intelligent Systems
Knowledge base created using simulation of a large number cases and
system measurements
Automatic learning, data mining, and decision trees to build intelligent
systems
Fast analysis using a broad knowledge base and automatic decision making
Provides new insight into factors and system parameters affecting stability
More effective in dealing with uncertainties and large dimensioned problems
We just completed a PRECARN project: "POSSIT"
1529pk - 145
Copyright P. Kundur
DSA Using Intelligent Systems
1529pk - 146
Copyright P. Kundur
Real-Time Monitoring and Control:
An Emerging Technology
Advances in communications technology have made it possible to
monitor power systems over a wide area
remotely control many functions
Research on use of multisensor data fusion technology
process data from different monitors, integrate and process information
identify phenomenon associated with impending emergency
make intelligent control decisions
A fast and effective way to predict onset of emergency conditions and
take remedial actions
The ultimate "self-healing" power system !
1529pk - 147
Copyright P. Kundur
Distributed Generation (DG)
Offer significant economic, environmental and security benefits
Microturbines
small, high speed power plants
operate on natural gas or gas from landfills
Fuel Cells
combines hydrogen with oxygen from air to generate electricity
hydrogen may be supplied from an external source or generated inside
fuel cell by reforming a hydrocarbon fuel
Not vulnerable to power grid failure due to system instability or
natural calamities
protection and controls should be designed so that units continue to
operate when isolated from the grid
1529pk - 148
Copyright P. Kundur
Reliability Management System
Roles and responsibilities of individual entities
well chosen, clearly defined and properly enforced
Coordination of reliability management
Need for a single entity with overall responsibility for security of
entire interconnected system
real-time decisions
System operators with high level of expertise in system stability
phenomena, tools
1529pk - 149
Copyright P. Kundur
Summary
1. The new electricity supply industry presents increasing challenges for
stable and secure operation of power systems
2. State-of-the-art methods have advanced our capabilities significantly
comprehensive stability analysis tools
automated tools for system planning/design
on-line Dynamic Security Assessment (DSA)
coordinated design of robust stability controls
3. Industry is yet to take full advantage of these developments
cont'd
1529pk - 150
Copyright P. Kundur
Summary (cont'd)
4. Future directions will be to explore new techniques which can
better deal with growing uncertainties and increasing
complexities of the problem
risk-based security assessment
intelligent systems for DSA
"self-healing" power systems
real-time monitoring and control
5. Wide-spread use of distributed generation could be a cost
effective means of minimizing the impact of power grid failures