An Automated Identification Approach for Partial Discharge Detection Using Density-Based Clustering Without User Inputs

Md Anisur Rahman; Li-Minn Ang; Kah Phooi Seng

doi:10.1109/TAI.2023.3237648

Back

An Automated Identification Approach for Partial Discharge Detection Using Density-Based Clustering Without User Inputs

Journal article

Peer reviewed

An Automated Identification Approach for Partial Discharge Detection Using Density-Based Clustering Without User Inputs

Md Anisur Rahman, Li-Minn Ang and Kah Phooi Seng

IEEE Transactions on Artificial Intelligence, Vol.5(1), pp.310-320

2024

DOI: https://doi.org/10.1109/TAI.2023.3237648

Files and links (1)

url

https://doi.org/10.1109/TAI.2023.3237648View

Published Version

Abstract

clustering

partial discharge

kernel density estimation

vein-based clustering

cluster merging

A recently published state-of-the-art density-based clustering technique called ICFSFDP for partial discharge detection requires various sensitive user-defined input parameters. This paper presents a parameter-free clustering technique called PDAutoClust for partial discharge detection. The PDAutoClust algorithm can produce high-quality clusters for partial discharge datasets without requiring user-defined input parameters. PDAutoClust produces high-quality clustering results by utilizing a vein-based density clustering approach. The vein of a cluster is produced by using multivariate kernel density estimation (KDE) and a unique neighborhood set (UNS). We compared the performance of PDAutoClust against ICFSFDP and seven other state-of-the-art density-based and non-density-based clustering techniques by using four partial discharge datasets in terms of adjusted rand index, normalized mutual information, F1-score, and purity. Another contribution of the paper is a novel merging technique used with PDAutoClust to merge small non-viable clusters that a clustering technique may produce. PDAutoClust produces the final clusters for a dataset by merging the non-viable clusters that a clustering technique may produce. We also evaluate the performance of PDAutoClust with merging technique versus PDAutoClust without merging technique using four datasets. Simulation results for PDAutoClust with the merging technique show good performance compared to ICFSFDP and seven other state-of-the-art clustering techniques. We also performed an ablation study to demonstrate the importance of the steps involved in PDAutoClust.

Details

Title: An Automated Identification Approach for Partial Discharge Detection Using Density-Based Clustering Without User Inputs
Authors: Md Anisur Rahman (Author) - Charles Sturt University
Li-Minn Ang (Author) - University of the Sunshine Coast, Queensland, School of Science, Technology and Engineering
Kah Phooi Seng (Author) - Queensland University of Technology
Publication details: IEEE Transactions on Artificial Intelligence, Vol.5(1), pp.310-320
Publisher: Institute of Electrical and Electronics Engineers
Date published: 2024
DOI: 10.1109/TAI.2023.3237648
ISSN: 2691-4581
Organisation Unit: University of the Sunshine Coast, Queensland; School of Science, Technology and Engineering; Engage Research Lab
Language: English
Record Identifier: 99701598502621
Output Type: Journal article

Metrics

109 Record Views