Aplikasi Backpropagation Neural Network untuk Deteksi Gangguan Sistem Tenaga Listrik Listrik pada Rele Jarak  

Oleh:

Saleh, Azmi

Jenis: Article from Journal - ilmiah nasional
Dalam koleksi: Jurnal Teknologi Industri vol. 7 no. 1 (Jan. 2003), page 19-30.
Topik: Distance Relay; Power System; Backpopagation Neural Network; Recognition

Ketersediaan

Perpustakaan Pusat (Semanggi) Nomor Panggil: JJ83.4 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

A distance relat for the transmission lines protection is usually designed on fixed setting range. If the impedance of the transmission lines to be protected is not on the impedance setting range of the distance relay, the relay could not be used. It is purposed the use of backpropagation neural network as a pattern recognizer for distance relay in fault detection by recognizing pattern of voltage an current wave forms. The principle of backpropagation neural network in distance relay is mapping to recognize voltage and current wave forms. Backpropagation neural network is trained supervisely by implementing Generalized Delta Rule algorithm (GDR) to recognize the pattern of voltage and current wave forms of the transmission line in a faulted condition by applying phase voltage and current as the inputs. The output of backpropagation neural network in this application is trip/no trip decision. If the mapping result in values between 0.5-1.0, the distance relay should send a trip signal to the CB. The distance relay should not send a trip signal result in values between 0.0-0.49.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali