Enhancing PHM System of Aircraft Generator with Machine Learning-Driven Faults Classification

Umar Saleem, Weinjie Liu, Weilin Li, M. U. Sardar, Muhammad Mobeen Aslam, Saleem Riaz
{"title":"Enhancing PHM System of Aircraft Generator with Machine Learning-Driven Faults Classification","authors":"Umar Saleem, Weinjie Liu, Weilin Li, M. U. Sardar, Muhammad Mobeen Aslam, Saleem Riaz","doi":"10.1109/ICETSIS61505.2024.10459418","DOIUrl":null,"url":null,"abstract":"Prognostic and Health Management (PHM) played a vital role in the industrial revolution. An efficient PHM system improves reliability and safety by detecting whether an industrial component has deviated from its normal operating condition, predicting when a fault will occur, and classifying the type of fault. Due to the rapid development of more electric aircraft in recent years, the electric power system of aircraft has become more critical in ensuring safe flying. This research mainly focuses on classifying aircraft generator faults using the Support Vector Machine (SVM). To use the SVM for fault classification, firstly, create a data set of 1112 records containing all possible types of short circuit faults and normal states using the MATLAB Simulink model. Extract features from these records by decomposing them with Wavelet Transform. The principal component analysis (PCA) optimization technique is used on detail coefficients for trained SVM that will correctly classify generator faults. Then, train the SVM at each type of fault and normal state using 70% of the data and test it on the remaining 30%. It has been observed that if the system works under normal working conditions, all SVM output will be zero. In the faulty condition, the SVM output that belongs to the type or class of fault will be one and will display the type of fault. The suggested technique has been extensively evaluated for several fault types under various operating conditions. The SVM results demonstrate impressive accuracy in fault classification and significantly improve aviation generators' PHM systems.","PeriodicalId":518932,"journal":{"name":"2024 ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems (ICETSIS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2024 ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems (ICETSIS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICETSIS61505.2024.10459418","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Prognostic and Health Management (PHM) played a vital role in the industrial revolution. An efficient PHM system improves reliability and safety by detecting whether an industrial component has deviated from its normal operating condition, predicting when a fault will occur, and classifying the type of fault. Due to the rapid development of more electric aircraft in recent years, the electric power system of aircraft has become more critical in ensuring safe flying. This research mainly focuses on classifying aircraft generator faults using the Support Vector Machine (SVM). To use the SVM for fault classification, firstly, create a data set of 1112 records containing all possible types of short circuit faults and normal states using the MATLAB Simulink model. Extract features from these records by decomposing them with Wavelet Transform. The principal component analysis (PCA) optimization technique is used on detail coefficients for trained SVM that will correctly classify generator faults. Then, train the SVM at each type of fault and normal state using 70% of the data and test it on the remaining 30%. It has been observed that if the system works under normal working conditions, all SVM output will be zero. In the faulty condition, the SVM output that belongs to the type or class of fault will be one and will display the type of fault. The suggested technique has been extensively evaluated for several fault types under various operating conditions. The SVM results demonstrate impressive accuracy in fault classification and significantly improve aviation generators' PHM systems.
利用机器学习驱动的故障分类改进飞机发电机 PHM 系统
诊断与健康管理(PHM)在工业革命中发挥着至关重要的作用。高效的 PHM 系统可以检测工业部件是否偏离正常工作状态、预测故障发生时间并对故障类型进行分类,从而提高可靠性和安全性。由于近年来电动飞机的快速发展,飞机的电力系统在确保飞行安全方面变得更加关键。本研究主要侧重于使用支持向量机(SVM)对飞机发电机故障进行分类。要使用 SVM 进行故障分类,首先要使用 MATLAB Simulink 模型创建一个包含所有可能的短路故障类型和正常状态的 1112 条记录的数据集。使用小波变换对这些记录进行分解,从中提取特征。对细节系数采用主成分分析 (PCA) 优化技术,训练出能正确分类发电机故障的 SVM。然后,使用 70% 的数据在每种故障类型和正常状态下训练 SVM,并在剩余的 30% 数据上进行测试。据观察,如果系统在正常工作条件下运行,所有 SVM 输出都将为零。在故障状态下,属于故障类型或类别的 SVM 输出将为 1,并显示故障类型。所建议的技术已在各种工作条件下针对几种故障类型进行了广泛评估。SVM 的结果表明,故障分类的准确性令人印象深刻,并极大地改进了航空发电机的 PHM 系统。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信