ANN based directional overcurrent relay

Author

Vishwakarma, D.N. ; Moravej, Zahra

Author_Institution

Dept. of Electr. Eng., Banaras Hindu Univ., Varanasi, India

Volume

1

fYear

2001

fDate

2001

Firstpage

59

Abstract

This paper describes the design and development of an ANN based novel directional overcurrent relay. A multi-layer feedforward neural network (FFNN) trained with the error backpropagation algorithm has been developed to realize this relay. The ANN based algorithm is off line tested using different simulation equations for overcurrent and voltage under fault conditions to evaluate the performance of the proposed method in terms of accuracy and speed. The proposed ANN based directional overcurrent relay shows promising dependability, security and speed of operation

Keywords

backpropagation; control system analysis; control system synthesis; feedforward neural nets; neurocontrollers; overcurrent protection; power system control; power system protection; power system relaying; relay protection; ANN-based directional overcurrent relay; design; development; error backpropagation algorithm; fault conditions; multi-layer feedforward neural network; overcurrent; overvoltage; simulation equations; Artificial neural networks; Backpropagation algorithms; Differential equations; Feedforward neural networks; Multi-layer neural network; Neural networks; Relays; Security; Testing; Voltage;

fLanguage

English

Publisher

ieee

Conference_Titel

Transmission and Distribution Conference and Exposition, 2001 IEEE/PES

Conference_Location

Atlanta, GA

Print_ISBN

0-7803-7285-9

Type

conf

DOI

10.1109/TDC.2001.971209

Filename

971209