This work presents an algorithm based on Artificial Neural Networks (ANN) for the implementation of a distance relay to be used in transmission lines. The main contribution of the proposed algorithm is the possibility to protect any transmission line (generic algorithm) independently of its voltage level or configuration, without a new ANN process training. Therefore, unlike previous works, this algorithm does not need any adjustment in the ANN architecture and its parameters, when applied to different electric systems. This characteristic is the main differential of the solution presented here, when compared to other ANN based algorithms, always developed to operate in only one particular transmission line (tower geometry, cable, length and pre-determined voltage level), which makes them unfit to be implemented in a commercial relays.
[1]
B. Kulicke,et al.
Multi neural network based fault area estimation for high speed protective relaying
,
1996
.
[2]
M. M. Morcos,et al.
ANN-based techniques for estimating fault location on transmission lines using Prony method
,
2001
.
[3]
Mladen Kezunovic,et al.
Detect and classify faults using neural nets
,
1996
.
[4]
Y. I. Li,et al.
An ANN-based distance protective relays of transmission lines
,
2001
.
[5]
Tahar Bouthiba,et al.
Fault location in EHV transmission lines using artificial neural networks
,
2004
.