Chapt 1

CHAPTER 1
INTRODUCTION
In the modern energy management systems, network application functions are used as
valuable tools to balance increasingly stringent operating conditions against economic efficiency.
With the growing stress on today’s power systems, many utilities increasingly face the threat of
transient stability problems. There is a pressing need for inclusion of on-line dynamic security
analysis capabilities in the energy management systems. Dynamic security analysis entails
evaluation of the ability of the power system to withstand a set of severe but credible
contingencies and to survive transition to an acceptable steady-state condition. Both transient
stability and voltage stability are often of concern in these systems.
Various methods for off-line transient stability analysis are under research and
development. They include efficient step-by-step integration of the differential equations, direct
methods of stability analysis, pattern recognition techniques, expert systems, and neural networks.
The step-by-step time simulation is a conventional method, but also the most reliable one we
present. However, efficient computation of a stability margin is needed to rank different
contingencies and indicate the remedial actions required to alleviate or reduce the threat of
instability. Among different methods proposed in the literature, the standard “Energy Approach”
has received a great deal of attention. It belongs to the class of direct methods for stability
analysis. A direct method for transient stability analysis is defined in [1] as a method that is able
to determine stability without explicitly integrating differential equations describing the post-fault
system. For example, it offers the opportunity of assessing the transient stability of power
systems more directly and effectively than the conventional approach based on simulation. More
fundamentally and, in terms of potential applications, more significantly, it also provides a
quantitative measure of how stable or unstable a particular case may be. However, direct
methods have one disadvantage when compared to conventional step-by-step methods. The
models used in direct methods are less detailed. For example, turbine governors, voltage
regulators and static VAR compensator are not considered. One of the main idea of the Energy
Approach as a direct method is that if the transient energy of a multimachine power system could
be measured fast enough and system corrections could be made soon enough, the system could
withstand a set of severe disturbance by injecting energy equal and opposite to the disturbance.
Due to major developments in the area of real-time phasor measurements (further

explanations on phasor measurements are in appendix A [20],[21]), it is now possible to measure
at a fast sampling rate the phase angles at the terminal buses of all the machines in a system.
1
Thanks to PMUs, a good and complete generator model is no longer needed. This overcome one
of the major drawback of the direct methods. Because of these real-time synchronized
measurement capabilities, a recent focus in power system research [1,6,12] has been on coming
up with a real-time way to assess the transient stability margins of a system following a major
disturbance in the system. To realize such task, a new computationally fast method is developed
in this thesis to assess the system transient stability margins in a range of a hundred of
milliseconds, such that proper control mechanisms can be activated in time to maintain stability.
This method is developed not only to overcome the transient stability assessment problems
but also to deal with voltage regulators, static VAR compensator or dynamic braking as control
tools [19]. However, this thesis will mainly focus on predicting the trajectory in the angle space
and assessing the transient stability margin in real-time. The last chapter outlines how to integrate
control tools. Those directions have to be considered as proposals, and that much more research
will be necessary to implement them. Successfully tested on 3 and 4 machine system and also on
the 10-machine New-England Power system, the developed algorithm is able to quickly enhance
the stability margin of the system at the first swing and also at the second swing. To complete the
prediction step, a second-order linear model and a second-order auto-regressive model are
estimated using a least-squares estimator. The method proposed by Y.Ohura, extended to fit an
iterative algorithm, is also computed. Then, these methods are compared with a time domain
simulation method that consists in integrating the dynamic equations using a fourth order Runga-
Kutta method. Also, the way to extend the prediction to subsequent swings is shown.
The prediction is stopped when the exit point is reached. This point is located at the
intersection of the predicted path and the Potential Energy Boundary Surface (p.e.b.s.) and used
as starting point to locate the c.u.e.p. The exit point is determined thanks to a combination of the
so called “ Ball-Drop” method [23] and the local maximum of the potential energy along a ray
starting at the Stable Equilibrium Point (s.e.p.) and the last point of the prediction. The c.u.e.p is
then computed using an improved version of the Shadowing method which makes use of the Ball-
drop method [22]. The transient potential energy at the c.u.e.p is compared to the transient
energy at the clearing time. The Energy Margin (EM) between these two values is used as index
of stability. If the EM is negative, the system is unstable; otherwise it is stable.
This thesis has been divided into six chapters. In the second chapter, the basic
multimachine stability theory, the different dynamic, state and transient energy equations are
presented. All the formulations are based on a Center of Angle notation. The theory is first
explained using one machine connected to an infinite-bus system and then extended to a
multimachine system. To illustrate the latter case, Athay’s 3-Machine System is used. In chapter
3, the different assumptions on which is based the proposed method are described. A detailed
methodology is also proposed and compared to a method proposed by Y.Ohura et al [6], a
method based on decision trees[5] and a method proposed by Bettiol et al [12]. Simulations have
been carried out on the 10-machine New-England power system. In chapter 4, various prediction
methods are described, their respective performances are compared and the computation of the
location of the exit point is explained. Chapter 5 focuses mainly on the computation of the c.u.e.p
2
once the exit point has been found. The approach consists in integrating the differential equations
of the reduced system as the BCU [3], [14] and the Shadowing methods [4] do. Chapter 6 gives
some conclusions on the proposed method and includes directions for further research.

Chapt 1

Uploaded by

Copyright:

Available Formats

Chapt 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapt 1

Uploaded by

Copyright:

Available Formats

CHAPTER 1

Due to major developments in the area of real-time phasor measurements (further

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.