Search In this Thesis
   Search In this Thesis  
العنوان
Advanced Signal Processing Techniques For Power System Protection \
المؤلف
Shaheen, Ayman El-Sayed Ibrahim.
هيئة الاعداد
باحث / AYMAN EL-SAYED IBRAHIM SHAHEEN
مشرف / Abdel-Maksoud I. Taalab
مشرف / Tamer A. Kawady
مشرف / Mohamed Abdel Maksoud Izzularab
الموضوع
Electric Power Systems - Protection. Electric Power Systems - Control. Signal Processing - Digital Techniques. Spread Spectrum Communications.
تاريخ النشر
2009.
عدد الصفحات
183 p. :
اللغة
الإنجليزية
الدرجة
الدكتوراه
التخصص
الهندسة الكهربائية والالكترونية
تاريخ الإجازة
1/1/2009
مكان الإجازة
جامعة المنوفية - كلية العلوم - ELECTRICAL ENGINEERING
الفهرس
Only 14 pages are availabe for public view

from 198

from 198

Abstract

During fault instants the measured power system signals usually have random behavior and contain high frequency components of short duration of time. The complicated nature of these signals and the substantial variation of their characteristics, accurate detection of features of such signals requires joint analysis in time and frequency domains. The unique properties of the Gabor Transform (GT), which include the distinctive optimal localization in the time frequency domains, promotes the idea of applying such transform for power system signal analysis and potentially helps in detection, accurate estimation and analysis of short time transient phenomena encountered during power system abnormalities and fault instants. The appropriate selection of the suitable signal processing technique plays an essential role for accessing the behavior of the corresponding modern protection system. In this work, the utilization of GT, as an advanced signal processing technique, for power system protection applications is presented. Different applications for transmission system protection are proposed covering detection, classification and location purposes. The proposed algorithms employ GT for features extraction instead of the conventional Fourier Transform (FT) analysis based methods and Neural Network (NN) for pattern recognition and classification processes. Moreover, complicated situations such as arcing faults are tested in order to investigate the behavior of the proposed scheme in such conditions. The contribution of this transform together with the neural network is evaluated through various simulation testes. The Alternative Transient Program (ATP) is employed for the investigation phase. A practical implementation of real-time application of the GT coefficients extraction using on-line training of NN is developed. The proposed scheme is implemented experimentally employing an advanced Digital Signal Processing ”DSP” from Texas Instruments part number ”TMS320C6416” operating at frequency of 720 MHz. The Spectrum Digital Inc development board TMS320C6416 DSK was utilized for such purpose. This aims to confirm the applicability of such scheme for real time applications of power system protection.