Acoustic emission partial discharge localization in oil based on artificial bee colony


Citation

Zhi, Yang Lim and Norhafiz Azis, . and Ahmad Hafiz Mohd Hashim, . and Mohd Amran Mohd Radzi, . and Nor Mohd Haziq Norsahperi, . and Azrul Mohd Ariffin, . (2025) Acoustic emission partial discharge localization in oil based on artificial bee colony. Pertanika Journal of Science & Technology (Malaysia), 33 (1). 241 -259. ISSN 2231-8526

Abstract

This study explores the application of an artificial bee colony (ABC) to locate partial discharge (PD) in a test tank based on acoustic emission (AE) approach. Data from a previous AE PD experimental study, which includes the coordinates of 3 AE sensors and the time difference of arrival (TDOA), were used to construct the nonlinear localization equations. It is known that localization algorithms are among the factors that can affect PD localization accuracy, and the ongoing research in this area underscores the need for further advancements in this topic. Therefore, the ABC was proposed to estimate the PD location through a colony of 120 bees, evenly divided into 60 employed and 60 onlooker bees. The employed bees explored the bounded search space, and onlooker bees refined PD locations found by the employed bees through local search. Scout bees were set out whenever a bee exceeded the limit of abandonment to discover possible PD locations in new areas of the search space. After 500 iterations, the optimal solution was the estimated PD location produced by ABC. Comparisons with the genetic algorithm (GA), particle swarm optimization (PSO) and bat algorithm (BA) revealed that the distance error, maximum deviation and computation time for AE PD localization based on ABC are the lowest. The study concludes that the ABC is more suitable for the multi-variable PD localization task than the GA, PSO, and BA due to its effective balance between local search by onlooker bees and global exploration by scout bees.


Download File

[img] Text (PJST Vol. 33 (1) Jan. 2025)
11 JST-5168-2024.pdf.crdownload

Download (1MB)

Abstract

This study explores the application of an artificial bee colony (ABC) to locate partial discharge (PD) in a test tank based on acoustic emission (AE) approach. Data from a previous AE PD experimental study, which includes the coordinates of 3 AE sensors and the time difference of arrival (TDOA), were used to construct the nonlinear localization equations. It is known that localization algorithms are among the factors that can affect PD localization accuracy, and the ongoing research in this area underscores the need for further advancements in this topic. Therefore, the ABC was proposed to estimate the PD location through a colony of 120 bees, evenly divided into 60 employed and 60 onlooker bees. The employed bees explored the bounded search space, and onlooker bees refined PD locations found by the employed bees through local search. Scout bees were set out whenever a bee exceeded the limit of abandonment to discover possible PD locations in new areas of the search space. After 500 iterations, the optimal solution was the estimated PD location produced by ABC. Comparisons with the genetic algorithm (GA), particle swarm optimization (PSO) and bat algorithm (BA) revealed that the distance error, maximum deviation and computation time for AE PD localization based on ABC are the lowest. The study concludes that the ABC is more suitable for the multi-variable PD localization task than the GA, PSO, and BA due to its effective balance between local search by onlooker bees and global exploration by scout bees.

Additional Metadata

[error in script]
Item Type: Article
AGROVOC Term: acoustic emission
AGROVOC Term: Apidae
AGROVOC Term: optimization methods
AGROVOC Term: data analysis
AGROVOC Term: research
AGROVOC Term: simulation models
AGROVOC Term: sensors
AGROVOC Term: accuracy
Geographical Term: Malaysia
Uncontrolled Keywords: Acoustic emission, artificial bee colony, localization, partial discharge, time difference of arrival
Depositing User: Ms. Azariah Hashim
Date Deposited: 17 Aug 2026 03:55
Last Modified: 17 Aug 2026 03:55
URI: http://webagris.upm.edu.my/id/eprint/4353

Actions (login required)

View Item View Item