A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine

Abstract The scarcity of samples and disunity of feature inputs hinder the enhancement of transformer fault diagnosis performance, and there are mutual influences between model construction and feature selection, which cannot only consider a single process. Therefore, this study proposes a novel two...

Full description

Saved in:
Bibliographic Details
Main Authors: Hanyu Shi, Mingxia Chen
Format: Article
Language:English
Published: Wiley 2023-03-01
Series:IET Electric Power Applications
Subjects:
Online Access:https://doi.org/10.1049/elp2.12270
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722266739277824
author Hanyu Shi
Mingxia Chen
author_facet Hanyu Shi
Mingxia Chen
author_sort Hanyu Shi
collection DOAJ
description Abstract The scarcity of samples and disunity of feature inputs hinder the enhancement of transformer fault diagnosis performance, and there are mutual influences between model construction and feature selection, which cannot only consider a single process. Therefore, this study proposes a novel two‐stage transformer fault diagnosis strategy, which includes a multi‐filter interactive feature selection method (MIFS) constructed, and a diagnosis model ASSA‐SVM based on the adaptive sparrow algorithm (ASSA) optimised support vector machine (SVM). Firstly, the proposed MIFS incorporates ReliefF and mRMR to establish a comprehensive criterion ReliefF‐mRMR for feature importance ranking, and then performs dimension‐by‐dimension input classifier interaction selection based on the ranking results to obtain the optimal feature subset. Secondly, ASSA was proposed to optimise the kernel parameters of SVM. A two‐stage integration model MIFS‐ASSA‐SVM was developed. Finally, Experiments were conducted using real fault data, and the diagnostic performance of different feature inputs, optimisation algorithms and classifiers were compared. The results show that the proposed method performs well on parameter optimisation, can dynamically and interactively select feature subsets with few dimensions and good generalisation performance, its overall diagnosis accuracy reached 92.47%, and the diagnosis performance of each fault type has good performance in multiple evaluation metrics.
format Article
id doaj-art-e874b19b03244ed7a3e48d8fca6ff44a
institution DOAJ
issn 1751-8660
1751-8679
language English
publishDate 2023-03-01
publisher Wiley
record_format Article
series IET Electric Power Applications
spelling doaj-art-e874b19b03244ed7a3e48d8fca6ff44a2025-08-20T03:11:24ZengWileyIET Electric Power Applications1751-86601751-86792023-03-0117334135710.1049/elp2.12270A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machineHanyu Shi0Mingxia Chen1School of Mechanical and Control Engineering Guilin University of Technology Guilin Guangxi ChinaSchool of Mechanical and Control Engineering Guilin University of Technology Guilin Guangxi ChinaAbstract The scarcity of samples and disunity of feature inputs hinder the enhancement of transformer fault diagnosis performance, and there are mutual influences between model construction and feature selection, which cannot only consider a single process. Therefore, this study proposes a novel two‐stage transformer fault diagnosis strategy, which includes a multi‐filter interactive feature selection method (MIFS) constructed, and a diagnosis model ASSA‐SVM based on the adaptive sparrow algorithm (ASSA) optimised support vector machine (SVM). Firstly, the proposed MIFS incorporates ReliefF and mRMR to establish a comprehensive criterion ReliefF‐mRMR for feature importance ranking, and then performs dimension‐by‐dimension input classifier interaction selection based on the ranking results to obtain the optimal feature subset. Secondly, ASSA was proposed to optimise the kernel parameters of SVM. A two‐stage integration model MIFS‐ASSA‐SVM was developed. Finally, Experiments were conducted using real fault data, and the diagnostic performance of different feature inputs, optimisation algorithms and classifiers were compared. The results show that the proposed method performs well on parameter optimisation, can dynamically and interactively select feature subsets with few dimensions and good generalisation performance, its overall diagnosis accuracy reached 92.47%, and the diagnosis performance of each fault type has good performance in multiple evaluation metrics.https://doi.org/10.1049/elp2.12270feature selectionsparrow algorithmsupport vector machinetransformer fault diagnosis
spellingShingle Hanyu Shi
Mingxia Chen
A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
IET Electric Power Applications
feature selection
sparrow algorithm
support vector machine
transformer fault diagnosis
title A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
title_full A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
title_fullStr A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
title_full_unstemmed A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
title_short A two‐stage transformer fault diagnosis method based multi‐filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
title_sort two stage transformer fault diagnosis method based multi filter interactive feature selection integrated adaptive sparrow algorithm optimised support vector machine
topic feature selection
sparrow algorithm
support vector machine
transformer fault diagnosis
url https://doi.org/10.1049/elp2.12270
work_keys_str_mv AT hanyushi atwostagetransformerfaultdiagnosismethodbasedmultifilterinteractivefeatureselectionintegratedadaptivesparrowalgorithmoptimisedsupportvectormachine
AT mingxiachen atwostagetransformerfaultdiagnosismethodbasedmultifilterinteractivefeatureselectionintegratedadaptivesparrowalgorithmoptimisedsupportvectormachine
AT hanyushi twostagetransformerfaultdiagnosismethodbasedmultifilterinteractivefeatureselectionintegratedadaptivesparrowalgorithmoptimisedsupportvectormachine
AT mingxiachen twostagetransformerfaultdiagnosismethodbasedmultifilterinteractivefeatureselectionintegratedadaptivesparrowalgorithmoptimisedsupportvectormachine