Detection of Fault on Transmission Line using Wavelet Transform and Neural Network Backpropagation

Em, Ennyvathana (2025) Detection of Fault on Transmission Line using Wavelet Transform and Neural Network Backpropagation. Masters thesis, institut Teknologi Sepuluh Nopember.

Text 6022231106-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (9MB) | Request a copy

Abstract

Reliability fault detection plays an essential role in maintaining the stability of power systems, particularly given that various faults originate from environmental influences or electrical failures. Although the Discrete Wavelet Transform (DWT) and Backpropagation Neural Network (BPNN) have been widely explored, the majority of existing studies rely on standard IEEE test systems. This study takes a different approach by applying these methods to the modified Kirirum 1 and Kirirum 3 transmission lines in Kampong Speu Province, Cambodia. The methodology involves applying DWT to extract features from voltage and current signal waveforms, followed by the application of BPNN for both fault classification and localization. The proposed model achieved an accuracy rate of 98.33%, with a fault localization error below 10% and an average detection time of 1 second across 960 simulated fault scenarios with resistance levels of 10 Ω, 20 Ω, 30 Ω, and 50 Ω, and different current flow capacity of 100%, 75%, 50%, 25% and 10%. These results demonstrate the method’s strong potential for application in localized power networks.
=================================================================================================================================
Deteksi gangguan keandalan memainkan peran penting dalam menjaga stabilitas sistem tenaga listrik, terutama mengingat berbagai gangguan berasal dari pengaruh lingkungan atau kegagalan listrik. Meskipun Transformasi Wavelet Diskrit (DWT) dan Jaringan Syaraf Tiruan Perambatan Balik (BPNN) telah banyak dieksplorasi, sebagian besar studi yang ada masih mengandalkan sistem uji IEEE standar. Studi ini mengambil pendekatan berbeda dengan menerapkan metode tersebut pada saluran transmisi Kirirum 1 dan Kirirum 3 yang dimodifikasi di Provinsi Kampong Speu, Kamboja. Metodologi yang digunakan meliputi penerapan DWT untuk mengekstraksi fitur dari bentuk gelombang sinyal tegangan dan arus, diikuti dengan penerapan BPNN untuk klasifikasi dan lokalisasi gangguan. Model yang diusulkan mencapai tingkat akurasi 98,33%, dengan kesalahan lokalisasi gangguan di bawah 10% dan waktu deteksi rata-rata 1 detik pada 960 skenario gangguan yang disimulasikan dengan tingkat resistansi 10 Ω, 20 Ω, 30 Ω, dan 50 Ω, serta kapasitas aliran arus yang berbeda, yaitu 100%, 75%, 50%, 25%, dan 10%. Hasil ini menunjukkan potensi kuat metode ini untuk diaplikasikan pada jaringan listrik terlokalisasi.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Discrete Wavelet Transform (DWT), Back-Propagation Neural Network (BPNN), Fault Detection, Fault Classification, Fault Location, Transmission Line.
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK201 Electric Power Transmission T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3226 Transients (Electricity). Electric power systems. Harmonics (Electric waves). T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105 Data Transmission Systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Enny Vathana Em
Date Deposited:	21 Jul 2025 01:09
Last Modified:	21 Jul 2025 01:09
URI:	http://repository.its.ac.id/id/eprint/120148

Actions (login required)

View Item