A Comparative Study and Improved Bearing Fault Classifier Using Raw Vibration Data Under Limited Training Samples

IF 1 Q4 ENGINEERING, MECHANICAL
J.S. Yap, M. Lim, M. S. Leong
{"title":"A Comparative Study and Improved Bearing Fault Classifier Using Raw Vibration Data Under Limited Training Samples","authors":"J.S. Yap, M. Lim, M. S. Leong","doi":"10.15282/ijame.21.1.2024.11.0857","DOIUrl":null,"url":null,"abstract":"Artificial intelligence is gaining traction in bearing fault detection and diagnosis. Generally, signal processing and feature selection are carried out to facilitate the fault classification process; however, classification accuracy tends to degrade under limited training data. In this paper, various artificial intelligence (AI) classification models are studied and compared using raw vibration data without signal processing and feature engineering. A Cosine k-Nearest Neighbours (CosKNN)-based classification model is optimized by integrating a Segmentive Mechanism, resulting in an overall classification F1-score improvement to 90.8% compared to the original classifier's 76.9%. The comparative findings show that the proposed model is suitable for circumstances with limited availability of training data, signal processing tools, and feature engineering tuning.","PeriodicalId":13935,"journal":{"name":"International Journal of Automotive and Mechanical Engineering","volume":null,"pages":null},"PeriodicalIF":1.0000,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Automotive and Mechanical Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.15282/ijame.21.1.2024.11.0857","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, MECHANICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Artificial intelligence is gaining traction in bearing fault detection and diagnosis. Generally, signal processing and feature selection are carried out to facilitate the fault classification process; however, classification accuracy tends to degrade under limited training data. In this paper, various artificial intelligence (AI) classification models are studied and compared using raw vibration data without signal processing and feature engineering. A Cosine k-Nearest Neighbours (CosKNN)-based classification model is optimized by integrating a Segmentive Mechanism, resulting in an overall classification F1-score improvement to 90.8% compared to the original classifier's 76.9%. The comparative findings show that the proposed model is suitable for circumstances with limited availability of training data, signal processing tools, and feature engineering tuning.
在有限的训练样本下使用原始振动数据进行比较研究和改进轴承故障分类器
人工智能在轴承故障检测和诊断中的应用越来越广泛。一般来说,进行信号处理和特征选择是为了促进故障分类过程;然而,在有限的训练数据下,分类精度往往会降低。本文使用原始振动数据,在不进行信号处理和特征工程的情况下,对各种人工智能(AI)分类模型进行了研究和比较。通过整合分段机制,优化了基于余弦 k 最近邻域(CosKNN)的分类模型,与原始分类器的 76.9% 相比,整体分类 F1 分数提高到 90.8%。比较结果表明,建议的模型适用于训练数据、信号处理工具和特征工程调整有限的情况。
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
CiteScore
2.40
自引率
10.00%
发文量
43
审稿时长
20 weeks
期刊介绍: The IJAME provides the forum for high-quality research communications and addresses all aspects of original experimental information based on theory and their applications. This journal welcomes all contributions from those who wish to report on new developments in automotive and mechanical engineering fields within the following scopes. -Engine/Emission Technology Automobile Body and Safety- Vehicle Dynamics- Automotive Electronics- Alternative Energy- Energy Conversion- Fuels and Lubricants - Combustion and Reacting Flows- New and Renewable Energy Technologies- Automotive Electrical Systems- Automotive Materials- Automotive Transmission- Automotive Pollution and Control- Vehicle Maintenance- Intelligent Vehicle/Transportation Systems- Fuel Cell, Hybrid, Electrical Vehicle and Other Fields of Automotive Engineering- Engineering Management /TQM- Heat and Mass Transfer- Fluid and Thermal Engineering- CAE/FEA/CAD/CFD- Engineering Mechanics- Modeling and Simulation- Metallurgy/ Materials Engineering- Applied Mechanics- Thermodynamics- Agricultural Machinery and Equipment- Mechatronics- Automatic Control- Multidisciplinary design and optimization - Fluid Mechanics and Dynamics- Thermal-Fluids Machinery- Experimental and Computational Mechanics - Measurement and Instrumentation- HVAC- Manufacturing Systems- Materials Processing- Noise and Vibration- Composite and Polymer Materials- Biomechanical Engineering- Fatigue and Fracture Mechanics- Machine Components design- Gas Turbine- Power Plant Engineering- Artificial Intelligent/Neural Network- Robotic Systems- Solar Energy- Powder Metallurgy and Metal Ceramics- Discrete Systems- Non-linear Analysis- Structural Analysis- Tribology- Engineering Materials- Mechanical Systems and Technology- Pneumatic and Hydraulic Systems - Failure Analysis- Any other related topics.
×
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