Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/4559
Title: | Intelligent systems techniques for electrical power systems fault location |
Researcher: | Chandra Sekhar, P |
Guide(s): | Sanker Ram, B V |
Keywords: | Power system Power Quality Problems High voltage spikes Transients Voltage sag magnitude Electrical faults Neuro computing Electrical Engineering |
Upload Date: | 5-Sep-2012 |
University: | Jawaharlal Nehru Technological University |
Completed Date: | March, 2012 |
Abstract: | Modern power system network consists of several equipments like synchronous generators, transformers, transmission lines, bus- bars, loads etc; operating at different capacities and voltage levels. As a result, the complexity of the power system network has been increased. Due to this the power system network is frequently subjected to disturbances like short circuits, faults.In this work an approach to minimize neural network training error has been developed. Even the minor disturbances are detected and classified accurately using neural networks and wavelet transforms. The developed algorithm is implemented on a system subjected to different fault conditions.In this work an algorithm is developed to overcome the disturbances and accurately detect and classify the disturbances in the signal. The developed algorithm is independent of load voltage and customized for several sampling frequencies. The algorithm is tested on a distributed network having 10 nodes for several fault conditions and the result obtained is accurate, thereby providing an efficient architecture for the development of fully automated monitoring systems with classification ability in distributed power system. Among the components in a power system network, transmission line experiences most of the disturbances as it is exposed to the environment. In this work a fully connected multilayer feed forward neural network is trained with a supervised learning algorithm known as Back Propagation Algorithm (BPA).The neural fault detectors and locators are trained and tested for several fault conditions (fault types, fault locations, fault resistances and fault inception angles) and various power system data (source capacities, source voltages, source angles, time constants of the sources in a selected network model.Further, fuzzy neural network architecture for digital fault processing in distributed network is developed. |
Pagination: | xvii, 155p. |
URI: | http://hdl.handle.net/10603/4559 |
Appears in Departments: | Department of Electrical and Electronics Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 76.25 kB | Adobe PDF | View/Open |
02_declaration.pdf | 88.03 kB | Adobe PDF | View/Open | |
03_certificate.pdf | 124.58 kB | Adobe PDF | View/Open | |
04_acknowledgements.pdf | 110.16 kB | Adobe PDF | View/Open | |
05_abstract.pdf | 72.71 kB | Adobe PDF | View/Open | |
06_contents.pdf | 91.92 kB | Adobe PDF | View/Open | |
07_list of figures.pdf | 183.59 kB | Adobe PDF | View/Open | |
08_list of tables.pdf | 78.19 kB | Adobe PDF | View/Open | |
09_list of symbols and abbreviations.pdf | 131.61 kB | Adobe PDF | View/Open | |
10_chapter 1.pdf | 165.81 kB | Adobe PDF | View/Open | |
11_chapter 2.pdf | 312.29 kB | Adobe PDF | View/Open | |
12_chapter 3.pdf | 1.41 MB | Adobe PDF | View/Open | |
13_chapter 4.pdf | 2.43 MB | Adobe PDF | View/Open | |
14_chapter 5.pdf | 75.71 kB | Adobe PDF | View/Open | |
15_references.pdf | 123.8 kB | Adobe PDF | View/Open | |
16_appendices.pdf | 176.49 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: