Jahanger, Hayder K., Kherif, Omar, Robson, Stephen ![]() ![]() |
Abstract
The fault location in distribution network with multiple branches and mixed overhead lines and underground cable is of vital importance. A hybrid approach that combines Artificial Neural Network (ANN) and travelling wave based double-ended method in distribution network with multiple branches and mixed lines is investigated. An ANN is trained using limited amount of measurement to identify the faulty section in four terminals distribution network. Then, a double-ended traveling wave fault location is applied using terminals choice based on the ANN decision. The training data is generated using ATP-EMTP simulation software and the data processed using CWT to extract Time of Arrival (ToA) and analysed using MATLAB script. The analysis showed that the ANN is capable to identify fault section with accuracy of 90% or more while the fault location accuracy is within 99%. The algorithm is resilience to fault resistance, inception angle and fault types while it is influenced by the system added noise. However, analysing the recorded signal with multiple CWT levels enable the correct extraction of the fault wave arrival time.
Item Type: | Conference or Workshop Item (Paper) |
---|---|
Date Type: | Publication |
Status: | Published |
Schools: | Engineering |
Publisher: | IEEE |
ISBN: | 9781837240852 |
Last Modified: | 02 Oct 2024 12:45 |
URI: | https://orca.cardiff.ac.uk/id/eprint/172431 |
Actions (repository staff only)
![]() |
Edit Item |