Assessment of transient stability of power systems using data from phasor measurement units

Number of pages: 122 File Format: word File Code: 32124
Year: 2013 University Degree: Master's degree Category: Electrical Engineering
  • Part of the Content
  • Contents & Resources
  • Summary of Assessment of transient stability of power systems using data from phasor measurement units

    Master thesis in the field of electrical-control engineering

    Abstract

    Evaluation of transient stability of power systems using data from phasor measurement units

     

    In effort

    Hania Mohammadi

    Rapid assessment of security in power networks in emergency situations and the occurrence of various errors is a vital thing to prevent collapse and create nationwide outages. Therefore, the evaluation of security in the power grid can have a preventive and effective control of the reliable and efficient operation of power grids all over the world. In static studies, the behavior of the system is examined in a permanent state, and the security situation in the power network has been investigated with a series of predictions. Since the amount of information received from large power networks is very large, we are trying to reduce the amount of information as much as possible by providing different feature selection methods such as correlation analysis or feature extraction such as principal component analysis. The reduced data are given as input to intelligent networks as a decision tree, and the evaluation of the security situation is done from these optimally trained trees.

    In dynamic security evaluation, after creating different working conditions, the behavior of the system is checked using the data received from the PMUs. These received data are processed in the field of time and frequency and are given as input to intelligent techniques such as decision trees and support vector machines to check the dynamic security of the power grid. In this approach, the effect of data reduction methods such as PCA has been investigated to create optimal and efficient SVM and DT. In addition, an idea for placing PMU with network visibility approach and also dynamic security assessment in power networks using decision tree and protective vector machines is presented. In this way, by entering the information of each bus bar individually or removing the information of that bus bar from the existing information of the network and checking the change of prediction error of the mentioned classifiers, the most important buses are selected to evaluate the security of network dynamics. The presented methods have been implemented on the sample network of 39 buses and the practical network of a part of Iran, and the results have been presented.

    Key words: phasor measurement unit, decision tree, support vector machines, static stability,

    Dynamic stability

    1-1- Statement of the problem

    In this thesis, we want to investigate various security issues, including static security and dynamic security, using the data received from PMUs. By using these data and smart techniques such as decision trees and support vector machines, the security situation in power networks is checked. Since the amount of information received from large power networks is very large, we are looking for solutions to reduce the amount of data as much as possible so that the reduced data contains a large part of the network information and the lost information can be ignored. By using feature selection and feature extraction techniques such as principal component analysis and correlation analysis, this dimension reduction is done. With this solution, SVM and DT inputs are reduced as much as possible and more optimal machine learning algorithms are produced suitable for real-time goals [1] and continuous updating. Also, with the approach of dynamic security assessment and also paying attention to the complete visibility of the network, a solution for optimal placement of PMUs using DT and SVM classifiers has been presented. In this way, the information of all buses are often used as representatives of the PMU installation for DT and SVM training, and according to the change of the error caused by the entry or exit of the information of each bus, an opinion is given regarding the best place to install the PMUs. It was important [1].  The first laboratory results of stability investigation were reported in 1924 [2] and the first results of stability investigation on a practical network were presented in 1925 [3]. An effective step in the development of steady state calculations was the development of the network analyzer in 1930. With the realization of drive systems with faster response, the transient instability in the first swing was limited and the limits of steady state power transfer were increased, but in some cases it caused a decrease in the damping of power swings, so the oscillatory instability was raised as a new problem. This process required more detailed modeling of synchronous machines and drive system. In the early 1950s, analog computers were used to analyze such problems. The first digital computer program for analyzing power system stability problems was presented in 1956. Most of the efforts and interests related to power system stability in the 1960s were devoted to transient stability. The result of these efforts was the creation of a powerful tool for transient stability analysis that was capable of analyzing very large networks and highly detailed models. Additionally, transient stability is significantly improved by using high-speed fault detection techniques and fast-response actuators and series compensators and special stability aids. New trends in the planning and operation of power systems have proposed new models of stability, which have caused fundamental changes in the dynamic characteristics of today's power networks. Unsustainable modes are becoming more complex day by day and require a comprehensive attention from all aspects of stability, so it is very necessary to adopt preventive control methods in these cases.[4] In the following, the history of classical methods and modern methods of stability investigation are presented separately.

    1-2-1 Classical methods:

    One ??of the methods of diagnosing transient stability is the use of time domain simulations of nonlinear differential equations of the power network, which was first proposed by Kundur. In this approach, step-by-step simulations must be performed in the time domain, which requires heavy calculations and requires detailed information about the network configuration during the occurrence of the error and after it, as a result, it is time-consuming and cannot be a suitable solution in real-time problems [4]. This issue is the main reason for the lack of system security assessment (DSA) [2] online and widely standard in the EMS package. [5]-[7].

    Methods based on transient energy functions have helped to perform security assessment directly without the need for time domain simulation [8]. In a solution proposed by Pai to detect stability after an event, the Transient Energy Function (TEF) is used based on Lyapanov stability, on the basis that the difference between kinetic and potential energy during the event and after the fault is cleared is calculated and compared with a predetermined critical value [9].

  • Contents & References of Assessment of transient stability of power systems using data from phasor measurement units

    List:

    Chapter One: Introduction

    1-1-Statement of the problem. 2

    1-2 research background. 3

    1-2-1 classic methods: 4

    1-2-2 modern methods using PMU data. 5

    1-3 research objective. 8

    1-4 The importance of research. 9

    1-5 chapters of the thesis. 10

    Chapter Two: Types of sustainability issues

    2- Types of sustainability issues. 13

    2-1 Sustainability classification criteria. 13

    2-2 Definition of static and dynamic stability. 13

    2-2-1 static stability (permanent) 13

    2-2-2 dynamic stability (transient) 14

    2-3 types of stability problems. 14

    2-3-1 rotor angular stability 14

    2-3-2 voltage stability 16

    2-3-3 frequency stability. 17

    Chapter Three: Static Voltage Security Assessment

    3-1 Statement of the problem. 21

    3-1-1 Collection of data required for static security assessment using PMU data. 22

    3-2 Introduction and training of decision tree: 24

    3-2-1 Decision tree: 25

    Title                                                                                                                                                                                                                                                           .

    3-3-1 methods based on feature extraction. 29

    3-3-1 method of principal component analysis or PCA. 30

    PCA algorithm. 32

    3-3-2 feature selection method using correlation analysis. 35

    3-4 proposed algorithm for rapid assessment of voltage security in power systems. 36

    3-4-1 Flowchart of static security evaluation algorithm using data received from PMUs 40

    3-5 Summary. 41

    Chapter 4: Evaluation of dynamic security in power networks

    4-Statement of the problem. 43

    4-1 Data collection for dynamic security assessment of the power grid. 43

    4-2- Introduction of decision-making indicators. 43

    4-2-1- COI signals. 44

    4-2-2- Features in the domain of time. 45

    4-2-3- Fast calculation of WASI in the frequency domain. 47

    4-2-4-Categorical Index 49

    4-3 vector of support machines. 50

    4-3-1 Structure of support vector machines (SVM). 51

    4-3-2 Design and training of support vector machines to evaluate the dynamic security of the system. 55

    4-4- Designing and training a decision tree to evaluate the dynamic security of the system. 56

    4-5 Optimal placement of PMUs with a dynamic security assessment approach and using intelligent techniques. 56

          4-5-1 Introduction of step-forward technique for placing PMU in the power grid. 57

    4-5-2 Introduction of step-back technique for placing PMU in the power grid. 58

    4-6 Examining the data volume reduction method (PCA) in dynamic security assessment of power system. 58

    Title Page

    Chapter 5: Simulation Results

    5-1- Introduction of the studied networks. 61

    5-2- Introduction of DIgSILENT emulator software. 62

    3-5 static studies of voltage in the 39-base model power network. 62

    5-3-1 Designing local decision trees for 39-base network. 63

    5-3-2 General decision tree training for 39-base network using dimensionality reduction techniques. 64

    Predictions 65

    5-3-3 General decision tree training for a part of Iran using data volume reduction techniques 68

    Chapter 5-4 of network dynamics studies 39 samples. 72

    5-4-1 Calculation of indices: 72

    5-4-2 Design and training of decision tree to evaluate dynamic security in 39-bus network. 73

    5-4-3 Design and training of support vector machines to evaluate dynamic security in 39-bus network. 77

    5-5 Using the data volume reduction method (PCA) in evaluating the security of the 39-base network. 81

    5-5-1 Using PCA and DT to evaluate the dynamic security of the 39 bus network. 81

    5-5-2 Using PCA and SVM to evaluate the dynamic security of the 39 bus network. 83

    5-5-3 The impact of PCA in reducing the effect of noise in the data received from PMUs 84

    5-6- PMU placement with the dynamic security assessment approach and using smart DT and SVM techniques. 85

    5-6-1 Placement of PMU using step forward and tree technique85

    5-6-1 Placement of PMU using step forward technique and decision tree. 86

    5-6-2 PMU placement using step forward technique and SVM. 88

    5-6-3 PMU placement using backward step technique and SVM. 89

    5-6-4 Placement of PMU using backward step technique and DT. 90

    5-7 dynamic security evaluation of the real network of southern Iran. 93

    5-8- Summary. 94

    Title

    Chapter 6: Conclusion and Suggestions

    6-1 Conclusion. 96

    6-2- Suggestions. 97

    List of references. 98

     

     

    Source:

     

    C. P. Steinmetz, “Power Control and Stability of Electric Generating Stations,” IEEE Trans on Power syst., vol. 5, pp. 1215, Dec. 1920.

    R. D. Evans and R. C. Bergvall, “Experimental Analysis of Stability and Power Limitations,” IEEE Trans on Power syst., pp. 39-58, 1924.

    R. Wilkins, "Practical aspects of system stability," IEEE Trans on Power syst., pp. 41-50, 1926.

    Prabha Kundur, Power System Stability And Control, Neal J. Balu And Mark G. Lauby, Ed. New York: McGraw-Hill, l, 1993. Y. Mansour, E. Vaahedi, A. Y. Chang, B. R. Corns, B. W. Garrett, K. Demaree, T. Athay, and K. Cheung, “B.C. Hydro's on-line transient stability assessment (TSA)—Model development, analysis, and post-processing,” IEEE Trans. Power Systems, vol. PWRS-10, no. 1, pp. 241–253, Feb. 1995.

    J. L. Jardim, C. A. da Neto, A. P. A. da Silva, D. M. Falcao, A. C. Z. de Souza, C. L. T. Borges, and G. N. Taranto, “A unified online security assessment system,” in CIGRE, Paris, 2000, pp. 38-102.

    V. Chadalavada, V. Vittal, G. C. Ejebe, G. D. Irisarri, J. Tong, G. Pieper, and M. McMullen, “An on-line contingency filtering scheme for dynamic security assessment,” IEEE Trans. Power Systems, vol. PWRS-12, no. 1, pp. 153–161, Feb. 1997.

    H. D. Chiang, C. S. Wang, and H. Li, “Development of BCU classifiers for on-line dynamic contingency screening of electric power systems,” IEEE Trans. Power Systems, vol. PWRS-14, no. 2, pp. 660–666, May 1999.

    M. A. Pai, Energy Function Analysis for Power System Stability. Boston, MA:Kluwer, 1989.

     Y.Xue, L.Wehenkel, R.Belhomme, P.Rousseaux, M.Pavella, Eeuxibie, B.heilbronn, and J.F.Lesigne, “Extended Equal Area Criterion,” IEEE Trans.Power Syst, vol. 7, no. 3, pp. 1012-1022, Aug 1992.

    D. Ruiz Vega and M. Pavella, "A comprehensive approach to transient stability control. I. Near optimal preventive control," IEEE Trans. Power Syst, vol. 18, no. 4, pp. 1446-1453, Nov 2003.

    V. Brandwajn, A. B. R. Kumar, A. Ipakchi, A. Bose, and S. D. Kuo, “Severity indices for contingency screening in dynamic security assessment,” IEEE Trans. Power Systems, vol. PWRS-12, no. 3, pp. 1136–1142, Aug. 1997. Y. Mansour, E. Vaahedi, M. A. El-Sharkawi, A. Y. Chang, B. R. Corns, and J. Tamby, “Large scale security screening and ranking using neural networks,” IEEE Trans. Power Systems, vol. PWRS-12, no. 2, pp.954–960, May 1997.

    F. Aboytes and R. Ramirez, “Transient stability assessment in longitudinal power systems using artificial neural network,” IEEE Trans. Power Systems, vol. PWRS-11, no. 4, pp. 2003–2010, Nov.  1996.

    G. C. Ejebe, C. Jing, J. G. Waight, V. Vittal, G. Pieper, F. Jamshidian, D. Sobajic, and P. Hirsch, “On-line dynamic security assessment: Transient energy based screening and monitoring for stability limits,” in 1997 IEEE/PES Summer Meeting, Berlin, Germany.

    M. La Scala, G. Lorusso, R. Sbrizzai, and M. Trovato, “A qualitative approach to transient stability analysis,” IEEE Trans. Power Systems, vol. PWRS-11, no. 4, pp. 1996–2002, Nov. 1996.

    M.Pavella, D.Ernst, and D.Ruiz-Vega, Transient Stability Of Power Systems. Boston, MA: Kluwer, 2000.

    A.G.Phadke, “Synchronized phasor measurements in power systems,” IEEE comput.Appl.Power, vol. 6, no. 2, pp. 10-15, Apr.

Assessment of transient stability of power systems using data from phasor measurement units