Object-Oriented Power System Transient Stability Modeling
AHMED GHERBI 1, MOHAMED BELKACEMI 2, BRUNO FRANCOIS 3
1 Laboratoryof Automaticsof Sétif (LAS), Department of Electrotechnics,
Facultyof Engineering, Universityof Setif, ALGERIA
2 Department of Electrotechnics, Faculty of Engineering,
University of Batna, ALGERIA
3L2EP, Ecole Centrale de Lille, Cité Scientifique, BP 48,
Villeneuve d'Ascq, 59651- FRANCE
Abstract: - The Unified Modeling Language (UML) has become part of the mainstream of object-oriented software development in a wide range of applications. This paper describes an efficient object-oriented software platform for power system transient stability modeling. This software may be used in educational purpose or in modeling, planning and analyzing of the power system with user-friendly Graphical User Interface (GUI) and object-oriented database. It consists on five major features: a simulation engine, a graphic editor for drawing the one line power system diagram which interacts with a visual database, graphic representations library and a set of applications e.g. the load flow, the transient stability, the fault analysis.
Key-Words: Power System, Simulation, Transient Stability, Object-Oriented Modeling, Windows-based.
1Introduction
Since the early decades of the century, electric power systems have been growing in size and complexity. The today's power engineers are faced with a variety of complex problems related with the planning, design and analyzing. The need of the use of efficient simulators in order to provide the required performance and reliability of current and future system has become necessary. To satisfy the continuous electric service, the power systems are designed to be stable under any disturbance [1-3].
Transient stability analysis (TSA) requires the evaluation of a power system’s ability to resist large disturbances and survive transitions to normal or acceptable operating conditions. Because of the severity and suddenness of the disturbance, the analysis of transient stability is focused on the first few cycles, following the fault occurrence or switching operation. Therefore, the study of power systems’ stability had long been recognized as one of the most important factors by both system planners and system operators [2,3]. The TSA is very complex since it involves modeling of the entire electric network as well as the control and the regulation systems associated with the generation units. It is modeled by a set of complex nonlinear algebraic and differential equations that can be simultaneously solved in time domain using the well known numerical integration methods [4,5].
It is well recognized that the Windows-based packages are very interesting in terms of the quick interpretation results and the interactive communication between the users and the computer processes. The Windows-based softwares are easier and simpler than in DOS environments and manage efficiency the main memory. The graphic representation of data and results make the physical interpretation of phenomena very comprehensive and understandable. The introduction of object-oriented languages using Unified Modeling Language (UML) has provided tools for developing several Windows-based packages that allow intuitive and visual interaction with the application programs [6-8].
In this article, making use of the object-oriented programming (OOP) in WindowsTM environment, we shall present the power system transient stability analysis modeling. This application has been incorporated as a separate module in developed Windows-based software called "Object-Oriented Power System Simulator" (OOPSS) [8,9]. It provides a friendly and easy to use tool in the transient stability analysis using a user-friendly Object-Oriented Graphical User Interface (OOGUI) environment and database system(OODB). The modelling part of OOPSS has been described in [8] using Object-Oriented Methodology (OOM) and UML. OOPSS and its modules have been implemented using Borland C++Builder-5.
2The Software Description
OOPSS uses the advantages of Windows-based programming. The Windows functions and features are included. It is structured in modular form in order to allow other packages to be superimposed on this structure to represent different proposed paradigms. It consists on the following major features:
- An Object Power System model (ObjectPS), which is the heart of the software [8].
- An Object Mathematical Library [8], which contains the main numerical methods for power system analysis using the benefits of OOM.
- A user-friendly OOGUI designed to visually edit the one-line diagram using component library.
- A visual Object-Oriented Database (OODB), which allows to the user to enter or modify the data of the components and the application parameters.
- A library of Object Oriented Power System Analysis Applications (OOPSAA) structured as independent modules. Each application has its own graphical and interactive interfacing window. In this sequel, we will describe the TSA module.
- Graphical User Interface
The main interaction between the user and the simulation is through a user friendly OOGUI whose most important component is the one-line diagrams. The OOGUI module provides the user with interactive visual communication with the solution process. It allows a better analysis and understanding the process of performing and studying power system. This module was designed using OOP techniques of encapsulation, inheritance and polymorphism [8,9]. The advantage of this technique includes flexibility for the user to interact virtually with all objects on the screen, small source and executable files, extensibility and reusability of existing code and a consistent GUI [7,10,11].
The communication between OOGUI and the computing models is done directly through OODB of each class. It means that every electrical object has its own graphic sample and its own data. The graphical editor consists on a component library and a canvas where the one-line diagram of power system is constructed. The component library contains a set of the commonly used graphical symbols of electrical equipments with their predefined parameters as default data.
The user builds the diagram by dragging symbols of electrical elements from the component library and releasing them at the appropriate position on the canvas. The connection between them is done through the use of the toolbars and arrows or the Windows dialog boxes. The components data are entered via dialog boxes at the same time as the diagram is drawn and may be modified at any moment the user wants as illustrated in Fig.1 for a 9-bus IEEE test power system [1]. The OOGUI environment allows some interesting features such that deleting, copying, cutting and pasting devices from the network diagram. When the diagram is finished, the network diagram must be first saved then compiled and debugged in order to verify the connectivity between components.
Fig.1General appearance of the graphical editor
2.2Object-Oriented Database
The numerical data of any component of the studied network is stored into a Windows-based database by entering or modifying the default data of each electrical device using Windows dialog boxes. These dialog boxes can be invoked by clicking on the graphic symbol of a device on the diagram. For instance, when the user double-clicks the generator symbol of the diagram, a dialog box is displayed on the screen then he can enter or modify the data using both the mouse and keyboard as shownin Fig.1. The entered data is stored into the corresponding equipment database as well as into the global OODB. Each dialog box contains "Help" button to allow the user to get the instructions about the components parameters. The database system has been incorporate in OOPSS as a separate object using OOP. Then, it can be invoked by any included individual application in the software. The Fig.2 shows the database of the studied power system. When "Ok" button is clicked, the bus admittance matrix of the system is automatically computed and stored. Since is very sparse, only the non-zero elements are calculated using Matrix class [8].
Fig.2Visual database management system
2.3Architecture Of OOPSAA Object
The software package contains a library of OOPSAA including Load-Flow (OOLF), Optimal-Power-Flow (OOOPF) [7], Fault analysis (OOFA), Transient Stability Analysis (OOTSA). All these applications use the information stored in the OODB which constitutes the core of the software.OOPSAAis built as an object that contains all possible object power system applications as instance classes OOPSAA. As shown in Fig.3, OOPSAA utilizes an instance class of Network [8,9] and communicates friendly with OOGUI and OODB.
Fig.3Object-Oriented Model of PSAA
3Transient Stability Modeling
Transient stability is one the most important application in power system design and operation. It has been included in OOPSS as an independent module. This one has been modeled as optimal power flow constrained. Therefore, it class declaration, called OOTSA class inherits its data and code from OOOPF class and only specific data and code are added to define this class as depicted in the following code fragment:
class OOTSA : public OOOPF {private:
int NbrEq ; //number of differential equations,
double Fonct(…);//function model
...... …...... … ...... …...... …
public:
vector<double> time;//integration vector time
double t01,t02; //initial time & clearing time
double tmax,Dt; //maximum iteration time...;
bool Switching; //prefault, faulted post-fault states
AnsiString ModelType, ...;// generator model type
void ElectricPower();//computes output electric power
void GenVoltage();//computes terminal d,q voltage
voidVoltRegulation();// perform voltage regulation
void SpeedRegulation(); //perform speedRegulation
voidExecute(); //Solve TSA Problem
//…. ……………Additional methods……………………
OOTSA (){new ...;...;}; // Constructor
~ OOTSA (){delete...;...;}; // Destructor
};
The method used for transient stability calculation consists of time-domain solution of a set of highly nonlinear differential equations representing the generator dynamics: subject to the initial conditions: and a set of nonlinear algebraic equations representing the network , where x is the vector of the state variables, also known as generator variables, which describe the dynamics of the system. The OOTSA communicates with ObjectMathLib and OODB/OOGUI modules and Power System Analysis Tools (called PSATools) object which aggregates some models (such as Pricing model) necessary for executing applications.
The problems related to transient stability have been the major area of attraction for power system planning and operation engineers all over the world. Various methods and techniques are used in transient stability analysis ranging from the traditional numerical methods to novel ones applying artificial intelligence algorithms e.g. Artificial Neural Network (ANN) [12] or Genetic Algorithm (GA). In this Windows-based software, and for education purpose, we have included the main used numerical integration methods (Euler, Modified Euler and Runge-Kutta) for the simultaneous solution of the set of the ordinary differential equations associated with the machines and their control system devices. Generating unit is modeled as follows:
a.Generator model
Several standard two-axis models (ranging from classical to 5th order) of the generator with field winding on the direct axis and damper winding on the quadrature axis [1,4,5]are considered for modeling transient stability problem. The 5th order model is usually considered to be satisfactorily precise to analyse electromechanical dynamics.State variable are phase angle, speed deviation, q-axis and d-axis component of internal voltage. The damping effect is included in the damping coefficient to maintain precision
b.Automatic voltage regulator model
A block diagram of the AVR model is shown in Fig. 4. the gains ,and and the time constants ,are adjusted to meet the original model.
Fig.4Block diagram of AVR model
c.Governor model
A block diagram of the governor turbine model, for thermal generators, is shown in Fig.5, where K is governor gain, KH is high pressure turbine load fraction, T3, T4 are gain constants and LVG is the low value gate.
Fig.5Block diagram of governor turbine model
4Transient stability input data
The user should enter the required input parameters values and select the options to solve the transient stability problem as depicted in Fig. 6. The regulators and machines parameters data are taken from the network database. The developed Windows-based program allows the user to select various models of regulations type, generators and loads as well as the fault data. The fault data includes the fault type (symmetrical or unsymmetrical), its location on the faulted line and the circuit breaker (CB) clearing times.
Once all the required parameters are entered, the transient stability analysis will process in time domain according the chosen start time and the end time for the iterative process using one numerical method among those included in the software.
Fig.6Input data for transient stability process
5Sample Results
The results of the transient stability are displayed in tabular and graphical form of the state variables at the specified time steps. The graphical results are displayed using different colors to allow comparison of the specified state variable for all generators. Fig.7 shows a sample resultof the generators rotor anglesevolution of the studied power system.
Fig.7Generatorsrotor angles evolution
6Conclusion
The complexity of the electric power system has been resulted in the need for a simulation and visualization tool for power system modeling, control and operation. This paper presents a Windows-based interactive and graphic software for education and training power system transient stability analysis.
The developed software satisfies the requirements of flexibility, extensibility, maintainability and data integrity. It can be useful in power system modeling, analysis and operation in user-friendly GUI. Using a developed component library, the user can quickly enter their power system in one-line diagram form using a graphical editor, without having to type cryptic data files in text format or entering parameter data in countless dialog boxes. A visual database system, with Windows dialog boxes for the data handling, has been proposed so that the user can manage the data related to power system. The developed package manages its memory requirement dynamically by itself using the features and benefits of Borland C++Builder 5.
References:
[1]P. M. Anderson, A. A. Fouad, Power System Control and Stability, IEEE Press, 1994.
[2]U.G. Knight, Power Systems in Emergencies- From Contingency planning to crisis Management,Edition Wiley, England, 2001,
[3]M. Crappe, Stabilité et Sauvegarde des Réseau Electriques, Ed. Hermès Science Pub.2003.
[4]G.W. Stagg, A.H. El-Abiad, Computer Methods in Power System Analysis, McGraw Hill, 1968.
[5]J. Arrillaga, C. P. Arnold, B. J. Harker, Computer Modeling of Electrical Power Systems, John Wiley & Sons, 1983.
[6]J.R Shin, W.H. Lee, D.H. Im, "A Windows–Based Interactive and Graphic Package for the Education and Training of Power System Analysis and Operation", IEEE Trans. On Power Systems, Vol. 14, No. 4, Nov. 1999.
[7]T. Bouktir, A. Gherbi, L. Belfarhi, M. Belkacemi, An Efficient Object-Oriented Load Flow Applied To A Large-Scale Power System: Application to Sonelgaz Network,Proc. ofInt. Conf. on Electrotechnics'2000, Oran,Algeria,
Proc. of ICEL'2000, pp. 443-447.
[8]A. Gherbi, M. Belkacemi, T. Bouktir,B. Francois, Object-Oriented Programming for Large Scale Power System Modelling”, (accepted paper), Conference Intern. de Génie Electrique (CIGE’04),2004, Setif, Algeria.
[9]A. Gherbi, M. Belkacemi, B. Francois, "An Object-Oriented Programming for Large Scale Power System Modelling”, 3rd Conference on Electrical Eng.(CEE’04), October 2004, Batna, Algeria
[10]D.C. Yu, H. Liu and F. Wu, "A GUI Based Visualization Tool for Sequence Networks", IEEE Trans. Power Syst, Vol. 13, No. 1, 1998.
[11]T.J. Overbye, P.W. Sauer, G. Gross, M.J. Laufenberg, J.D. Weber, "A Simulation Tool For Analysis Of Alternative Paradigms For The New Electricity Business", Proceedings Of The 20th Hawaii Int. Conf. On Syst. Sciences, Maui, H1, 1997, pp 634-640.
[12]A. Gherbi, S. Messalti, M. Belkacemi, S. Belkhiat, Energy Margin Assessment For Power System Transient Stability Using Neural Network,Conference Nationale sur l’Electrotechnique et ses Applications, 2004, Sidi Bel-Abbes, Algeria, Proc. ofCNEA’04, pp.193-198.