<abstract xmlns="http://www.w3.org/1999/xhtml"><p>Electricity distribution systems are subject to a variety of faults such as permanent and transient short circuits due to the extent and multiplicity of equipment. In principle, short circuit fault causes the existing protective equipment to operate and to no electricity the various parts of the distribution network. Rapid and accurate determination of fault location, repair and recovery, it has not prevented the distribution of energy. This will satisfy consumers and prevent the losses of electricity companies. In this paper, the artificial neural network and fault current profiles are used to determine the distance of the fault, determine the type of fault and detect the short circuit. This method provides the information needed to locate the fault by sampling the current before and after the fault occurs from the SCADA system. The effect of connectivity local resistance changes and the effect of load changes on fault location were evaluated. The results show that this method is more accurate than the voltage droop profile variation method in determining the fault distance and short circuit breakdown. If only the net fault current changes profile is used, the effect of the load changes in determining the short-circuit breakdown is much less.</p></abstract>

Electricity distribution systems are subject to a variety of faults such as permanent and transient short circuits due to the extent and multiplicity of equipment. In principle, short circuit fault causes the existing protective equipment to operate and to no electricity the various parts of the distribution network. Rapid and accurate determination of fault location, repair and recovery, it has not prevented the distribution of energy. This will satisfy consumers and prevent the losses of electricity companies. In this paper, the artificial neural network and fault current profiles are used to determine the distance of the fault, determine the type of fault and detect the short circuit. This method provides the information needed to locate the fault by sampling the current before and after the fault occurs from the SCADA system. The effect of connectivity local resistance changes and the effect of load changes on fault location were evaluated. The results show that this method is more accurate than the voltage droop profile variation method in determining the fault distance and short circuit breakdown. If only the net fault current changes profile is used, the effect of the load changes in determining the short-circuit breakdown is much less.

Fault Location in Radial Distribution Network Based on Fault Current Profile and the Artificial Neural Network

The Scientific Bulletin of Electrical Engineering Faculty

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Electricity distribution systems are subject to a variety of faults such as permanent and transient short circuits due to the extent and multiplicity of...

Fault Location in Radial Distribution Network Based on Fault Current Profile and the Artificial Neural Network

Data publikacji: 01 paź 2020

Zakres stron: 14 - 21

DOI: https://doi.org/10.2478/sbeef-2020-0103

Słowa kluczowe
Electrical distribution network, Fault location, Artificial neural network, Net fault current profile

© 2020 Majid Dashtdar et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Fault Location in Radial Distribution Network Based on Fault Current Profile and the Artificial Neural Network

Data publikacji: 01 paź 2020

Zakres stron: 14 - 21

DOI: https://doi.org/10.2478/sbeef-2020-0103

Słowa kluczoweElectrical distribution network, Fault location, Artificial neural network, Net fault current profile

© 2020 Majid Dashtdar et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Słowa kluczowe
Electrical distribution network, Fault location, Artificial neural network, Net fault current profile