Academic Journals Database
Disseminating quality controlled scientific knowledge

Algoritmo Neuro-Difuso para la Detección y Clasificación de Fallas en Líneas de Transmisión Eléctrica Usando ANFIS.

Author(s): Jhon A. Calderón | Germán Zapata M. | Demetrio A. Ovalle C.

Journal: Avances en Sistemas e Informática
ISSN 1657-7663

Volume: 4;
Issue: 1;
Start page: 101;
Date: 2007;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Fuzzy Neural Algorithm | Faults Detection and Classification: LIFs | HIFs | ANFIS (Adaptive Network-based Fuzzy Inference System)

The detection and classification of faults in electric transmission lines is an essential task to be performed. Within an electric power system a diversity of faults which come from low impedance faults (LIFs) to high impedance faults (HIFs) are exhibited. Last faults are more difficult to be detected due to the use of conventional distance relays. In addition, when they are not detected irreversible consequences are presented into the system. From above, it is inferred that taking into account that HIFs are less frequent than LIFs it appears essential to guaranty that any protection device must be able to satisfactorily detect both kind of electric faults. The aim of this paper is to present an algorithm to detect and classify both kind of faults LIFs and HIFs using ANFIS (Adaptive Network-based Fuzzy Inference System). The inputs for ANFIS are based on RMS (Root-Mean-Square) values from 3-phase and zero-sequence current. The obtained results show that an ANFIS model can detect and classify faults in a precise way including (LIFs y HIFs) in between a half cycle time.

Tango Rapperswil
Tango Rapperswil

RPA Switzerland

Robotic Process Automation Switzerland