Vision 2020 Dynamic Security Assessment in Real Time Environment
Vision 2020 Dynamic Security Assessment in Real Time Environment
Vision 2020 Dynamic Security Assessment in Real Time Environment
Vision 2020
Dynamic Security Assessment in Real Time
Environment
G. Bizjak, U. Kerin, University of Ljubljana, Slovenia
R. Krebs, E. Lerch, O. Ruhle, Siemens AG, Germany
©2008 IEEE.
2
and coordination of data processing. The DSA system is fully implementation and management of DSA tasks.
automated. DSA applications, including load flow, time The implementation is done by programming of syntaxes
domain simulation, and analysis, can be performed for a in common database, used as the reference by the simulation
number of cases (i) without user interaction. If security system. Syntaxes are grouped into macro instructions
constraints are violated, the DSA documents the contingency assessing areas of power system security. Macro specific
and more detailed investigation of a study case is possible. operations are defined using Block Oriented Simulation
Language (BOSL) and simulation system built-in functions.
A. Software Implementation
Simulation system used for implementation is
PSS™NETOMAC [8]. The simulation system provides open
environment for user interaction and enables simple
COMMON DATABASE
Reference network,
Events,
Next study case (i+) Study case Event
i Results,
(i) Etc.
Modular architecture
In computations, the macros are initialized in hierarchical developed user interface to create contingencies is shown in
order, so DSA associated tasks can be, in addition to the Fig. 4. The contingency builder allows selecting grid
horizontal management, also managed vertically on a elements, equipment and events and combination of them to
multilevel basis. Moreover, the use of batch definitions is write a scenario file.
possible. In general, DSA can be performed in two manners. The contingency screening process is implemented by
In a single manner one study case in a run is processed, Boolean algorithms. In processing, power system state
whereas in a continuous manner any number of study cases variables are searched for violations of limits representing the
can be processed. security criteria and if any are determined, Boolean variables
The common database comprises definitions of DSA adopt a binary TRUE (insecure) or in the opposite case a
associated tasks, including input/output data management, binary FALSE (secure). In final, each security issue is
security assessment procedures, simulation system control represented by a binary record, adequate for direct security
and monitoring functions. Some tasks are open for user assessment or for construction of security indices. An
modifications and can be modified by means of direct example of criteria and indices for contingency screening and
database access or by application of a graphical interface. In ranking is given in the following chapter.
either way, input data have to be in form of simulation system
readable code and inputted according to some input rules. The
4
k = 1,...., m
Ik ≤ Ik max (3)
are given in [5], [9]. ωi is the rotor speed with respect to COI
In [5], security indices based on ratio of maximum NG is the total number of generators
deviation achieved and maximal deviation admissible of Θi is the rotor angles with respect to COI
different state variables are defined. A ratio is calculated for a Θicl is the rotor angle of i-th generator at fault clearing
time frame considered and if it equals 1 or is greater the time
security violation is reported. The security indices given are
for rotor angle deviation of network generators, maximum and In the considered DSA system security indices have not
total frequency deviation, dynamic and quasi-stationary been implemented yet.
voltage deviation, power flow through transmission lines and C. Visualization and Monitoring
load shedding. An example of security index considering rotor
Clear graphical representation of power system security is
angle deviation is given by (9). For other indices see [5].
essential in recognizing weakest points of a system.
The Angle Index (AI) is defined as the minimum
Therefore, DSA must include meaningful visualization of
between 1 and maximum ratio of maximum deviation of the
information characterizing important security issues.
load angle of i-th generator and the maximum admissible load
Considered DSA system provides various output formats to
angle given by the protection relay. Namely, the relays,
meet these requirements, including interactive graphics,
protecting the generator against asynchronous operation, are
animations, matrix representation, electrical diagrams, tables,
adjusted is such a way that the load angle of a generator (δi)
etc. as shown in Fig. 5.
does not exceed a certain value (e.g. 120°). The value 1 of this
1) Interactive graphic
index represents that at least one generator in the system
Interactive graphic provides plots of system quantities with
reaches the considered limit.
time reference and enables output format manipulations. User
⎧⎪ ⎛ δ ci , max ⎞⎫ selected plots can be easily scaled by pre- or user-defined
AI = min ⎨1, max ⎜ ⎟⎪
⎬ (9) mathematical expressions and analyzed by application of
i =1,...NG ⎜ δ c , max, adm ⎟⎪
⎪⎩ ⎝ ⎠⎭ Fourier analysis. Moreover, each plot can be investigated in
detail by zooming in the time scope and using on-line cursor
In [9], a different approach is used. Introduced are the control.
indices based on three dot products. A dot product is defined
for detecting the exit point in the transient energy function. Detailed re-calculations Matrix-
The indices give the measure of total accelerating power and representation
the power system response to this accelerating power. The
indices are defined as follows:
Fehler! Es ist nicht möglich, durch die Bearbeitung von
Feldfunktionen Objekte zu erstellen.
∑ f ⋅ω
NG
dot1 = i i (10)
i =1 Animations
Numerical visualization &
general re-calculations
Mi
fi = Pmi − Pei − ⋅ PCOI (11)
Mt Figure 5 Visualization and monitoring of the dynamic behaviour of power
systems in case of system contingencies
∑ (P
NG
PCOI = − Pei ) (12) 2) Matrix representation
mi
i =1 Common approach in providing of security assessment
results is to use summary tables. These tables comprise a list
of investigated contingencies and information regarding
∑ f ⋅Θ
NG
dot 2 = i i (13) security violations. Matrix representation shown in Fig. 6
i =1 extends the visualization frame of such summary tables by
sorting the system events by applied contingency and related
∑ω ⋅(Θ − Θ )
NG network element, assessment criteria, and by organizing them
cl
dot 3 = (14)
i i i in reference to a colour scale. Moreover, the matrix comprises
i =1
hyperlinks enabling direct access to an event associated
where graphics.
Mi is the inertia constant of each generator
Mt is the total inertia constant of all generators
Pmi is the mechanical power input of each generator
Pei is the electrical power output for each generator
6
4) Other formats
In addition, other formats, such as for electrical
representation of a power system or for table representation of
security assessment associated data, are available. These
formats can be adjusted to meet user requirements and enable Figure 8. Interarea oscillation after300 MW trip, simulation results
multi-level overview of DSA results.
Using the eigenvalue mode of the DSA system the inter-
D. Study case area oscillations of the power system can be easily monitored,
The European UCTE system has been used to demonstrate moreover the system also shows how and which generators
the performance of the DSA. The system has an installed are involved in the oscillation (Fig. 9).
capacity of about 530 000 MW (2004) with a maximum load
demand of about 386 000 MW (2004). A model of the system
has been built with 610 generators, 4400 nodes, 12000 grid
branches, and 1050 controllers. The system model has been
validated using measurements of the installed Wide Area
Measuring System (WAMS).
Based on the comparison of the simulation results of a
300MW trip in Spain shown in Fig 8 and the recordings of
WAMS shown in Fig 7 it can be concluded that the model
represents the overall electromechanical system behaviour
with sufficient accuracy. The simulation has been performed
for 15 seconds under real time conditions. To achieve real
time conditions time steps of 10 ms are the limit. However,
for the electromechanically behaviour the accuracy with time
steps of 20 – 50 ms is also suitable.
measures can be checked very fast. Fig 10 depicts [6] M. Ni et al., ''Software implementation of online risk-based security
assessment'', IEEE transactions on power systems, Vol. 18, No. 3, pp.
countermeasures at different generators to increase damping 1165-1172, August 2003
in the system, here shown in the time domain, but analyzed in [7] Y. Xue et al., ''Quantitative assessment for transient voltage security'',
the frequency domain by system eigenvectors and residues. IEEE transactions on power systems, vol. 15, no. 3, pp. 1077-1083,
August 2000
[8] Lerch, E.; Kulicke, B.; Ruhle, O.; Winter, W.: “NETOMAC - Calculating,
Analyzing and Optimizing the Dynamic of Electrical Systems in Time and
Frequency Domain“, in Proc. 3rd IPST '99, Budapest, Ungarn, 20.-
24.06.1999
[9] Fu, C. and Bose, A. ''Contingency Ranking Based on Severity Indices in
Dynamic Security Analysis'', IEEE Transactions on Power Systems, Vol.
14. No. 3, pp. 980-986, August 1999
VII. BIOGRAPHIES