This doctoral thesis deals with the electric power systems (EPS) transient stability analysis.
Many approaches and methods have been developed over the years and important place
among them goes to the so-called direct methods. These methods are based on representing
the EPS using functions exhibiting special properties (i.e. Lyapunov function and transient
energy function). Direct methods allow one to assess the EPS stability level without
addressing the complex system of nonlinear differential equations which along with the
algebraic equations describe the EPS model. This thesis presents the application of direct
methods for transient stability analysis and implementation of EPS control strategies.
Approach to modelling and EPS dynamic simulation results evaluation is also presented.
In the first part of the thesis practical application of the direct EPS transient stability
assessment is presented. A software tool was developed for this purpose that allows one to
assess the EPS stability via direct method (we used the well-known PEBS method) within
a standard software for the EPS dynamic simulations. In this way already available
(standard) EPS models of an arbitrary size can be used. Besides, development of the basic
EPS simulation tools was avoided in this manner. The main advantage of the direct
approach is the possibility of forming a relatively fast recognition of the EPS transient
stability boundary. It should be noted, however, that the accuracy of the direct method is
slightly worse compared to the simulation method due to the certain fundamental
properties of the direct methods and the use of a simplified EPS model, which is a basis for
the transient energy function calculation. In the literature it is often pointed out that PEBS
method is in certain cases known to produce optimistic estimates of the stability region (i.e.
contingency which is determined as an unstable using the time-domain simulation appears
as stable using the PEBS method). Such cases appeared relatively seldom in the analysed
test systems. In a larger proportion of these cases the positive deviation of critical clearing
times was relatively small. The statistical analysis of the results also showed that the
assessment of the stability region was too conservative in some cases, which means that
further research and additional improvements of the developed method are required.
In addition to the EPS transient stability assessment, direct methods are also well
established in the field of control strategies for the non-linear systems, which also include
EPS. In this thesis we present a novel approach to improving the transient stability and
power oscillation damping using a multi-parametrical control device. The control strategy
is based on the use of an EPS energy function. The aim of the control is to achieve a rapid
dissipation of the energy gained during the disturbance. The energy-function time
derivative is expressed as a function of the generator’s electrical powers and the rotorangle
derivatives of the entire power system. In this way a globally optimum control
strategy, in the Lyapunov sense, is achieved. A characteristic of the energy-function time
derivative relative to the UPFC control parameters is numerically determined using an
online procedure during the dynamic simulation. Based on the determined characteristic
the UPFC parameters are controlled in such a way as to minimize the energy-function
derivative and achieve an efficient power evacuation during the generator’s first swing and
the power oscillation damping after the large disturbance in the EPS. The control strategy
was verified with computer simulations for Single-Machine-Infinite-Bus (SMIB) and
multi-machine test-system models, and the results are in line with the theoretical
considerations.
In this thesis a validation of a dynamic model of a Slovenian EPS using WAMS is also
presented. For this purpose a dynamic model of a part of the ENTSO-E system was
constructed by applying two professional dynamic-simulation software tools that use
different approaches to obtain mathematical solutions for the model to solve the problem.
The simulation results were compared with each other as well as with the measurements
gathered by the WAMS. Namely, Slovenia has almost full coverage of PMU
measurements at the high-voltage level (220- and 400-kV) buses. This fact has been used
to adapt the model, especially for the representation of the rest of the ENTSO-E network,
as well as for the validation of the results and explaining the physical background for the
existing differences. WAMS measurements for several different events occurring during
2010 and 2011 were used to adapt the model, which then exhibits a good match in all
situations recorded by the WAMS. The physical background for the resulting deviations is
presented. However, it should be pointed out that, after the model was finalized, for all the
validated cases the system element parameters remained unchanged. Of course, depending
on the simulated case, the initial steady-state conditions were set according to the
measurements.
In the last section of the thesis a field of an online dynamic security analysis (DSA) is
presented. Tools for an online DSA are very useful for the management of critical
situations in a transmission system operation. Online DSA has received a significant
attention in the last years. It should be designed to give the essential information about the
EPS stability during contingencies to the transmission system operator, whereby it is
essential that the stability assessment is carried out within the time frame that enables
taking appropriate measures which could prevent EPS reaching the stability region limit.
The online DSA was linked to the direct method by using the dynamic model of the
Slovenian EPS verified with the WAMS and analysed using the PEBS method as a
reference. Due to the variable accuracy of the stability region assessment, the applicability
of the results is limited, so it would be reasonable to apply the direct approach in a
combination with the time-domain simulation method (e.g. for the fast scan of potentially
dangerous contingencies).
|