Naslov
Artificial Neural Network Approach for Locating Faults in Power Transmission System

Autori
Teklić, Ljupko ; Filipović-Grčić, Božidar ; Pavičić, Ivan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
IEEE Eurocon 2013 Conference : proceedings / Kuzle, Igor ; Capuder, Tomislav ; Pandžić, Hrvoje - Zagreb : Institute of Electrical and Electronics Engineers (IEEE), 2013, 1425-1430

ISBN
978-1-4673-2231-7

Skup
IEEE Eurocon 2013 Conference

Mjesto i datum
Zagreb, Hrvatska, 01.07.2013. - 04.07.2013

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
fault location; transmission lines; feed forward neural network; artificial neural network

Sažetak
This paper presents fault location recognition in transmission power system using artificial neural network (ANN). Single phase short circuit on 110 kV transmission line fed from both ends was analyzed with various fault impedances, since it is the most common fault in power system. Load flow and short circuit calculations were performed with EMTP-RV software. Calculation results including currents and voltages at both line ends were used for training ANN in Matlab in order to obtain correct fault location and fault impedance, even for those cases that ANN has never encountered before. The network was trained with back propagation algorithm. Test results show that this approach provides robust and accurate location of faults for a variety of power system operating conditions and gives an accurate fault impedance assessment.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika

POVEZANOST RADA