用于机械故障诊断的新型自训练半监督深度学习方法

IF 7 2区 工程技术 Q1 ENGINEERING, INDUSTRIAL International Journal of Production Research Pub Date : 2023-12-02 DOI:10.1080/00207543.2022.2032860
Jianyu Long, Yibin Chen, Zhe Yang, Yunwei Huang, Chuan Li
{"title":"用于机械故障诊断的新型自训练半监督深度学习方法","authors":"Jianyu Long, Yibin Chen, Zhe Yang, Yunwei Huang, Chuan Li","doi":"10.1080/00207543.2022.2032860","DOIUrl":null,"url":null,"abstract":"Fault diagnosis is an indispensable basis for the collaborative maintenance in prognostic and health management. Most of existing data-driven fault diagnosis approaches are designed in the framework of supervised learning, which requires a large number of labelled samples. In this paper, a novel self-training semi-supervised deep learning (SSDL) approach is proposed to train a fault diagnosis model together with few labelled and abundant unlabelled samples. The addressed SSDL approach is realised by initialising a stacked sparse auto-encoder classifier using the labelled samples, and subsequently updating the classifier via sampling a few candidates with most reliable pseudo labels from the unlabelled samples step by step. Unlike the commonly used static sampling strategy in existing self-training semi-supervised frameworks, a gradually exploiting mechanism is proposed in SSDL to increase the number of selected pseudo-labelled candidates gradually. In addition, instead of using the prediction accuracy as the confidence estimation for pseudo-labels, a distance-based sampling criterion is designed to assign the label for each unlabelled sample by its nearest labelled sample based on their Euclidean distances in the deep feature space. The experimental results show that the proposed SSDL approach can achieve good prediction accuracy compared to other self-training semi-supervised learning algorithms.","PeriodicalId":14307,"journal":{"name":"International Journal of Production Research","volume":null,"pages":null},"PeriodicalIF":7.0000,"publicationDate":"2023-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"26","resultStr":"{\"title\":\"A novel self-training semi-supervised deep learning approach for machinery fault diagnosis\",\"authors\":\"Jianyu Long, Yibin Chen, Zhe Yang, Yunwei Huang, Chuan Li\",\"doi\":\"10.1080/00207543.2022.2032860\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Fault diagnosis is an indispensable basis for the collaborative maintenance in prognostic and health management. Most of existing data-driven fault diagnosis approaches are designed in the framework of supervised learning, which requires a large number of labelled samples. In this paper, a novel self-training semi-supervised deep learning (SSDL) approach is proposed to train a fault diagnosis model together with few labelled and abundant unlabelled samples. The addressed SSDL approach is realised by initialising a stacked sparse auto-encoder classifier using the labelled samples, and subsequently updating the classifier via sampling a few candidates with most reliable pseudo labels from the unlabelled samples step by step. Unlike the commonly used static sampling strategy in existing self-training semi-supervised frameworks, a gradually exploiting mechanism is proposed in SSDL to increase the number of selected pseudo-labelled candidates gradually. In addition, instead of using the prediction accuracy as the confidence estimation for pseudo-labels, a distance-based sampling criterion is designed to assign the label for each unlabelled sample by its nearest labelled sample based on their Euclidean distances in the deep feature space. The experimental results show that the proposed SSDL approach can achieve good prediction accuracy compared to other self-training semi-supervised learning algorithms.\",\"PeriodicalId\":14307,\"journal\":{\"name\":\"International Journal of Production Research\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":7.0000,\"publicationDate\":\"2023-12-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"26\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Production Research\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1080/00207543.2022.2032860\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, INDUSTRIAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Production Research","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1080/00207543.2022.2032860","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, INDUSTRIAL","Score":null,"Total":0}
引用次数: 26

摘要

故障诊断是预后和健康管理中协同维护不可缺少的基础。现有的大多数数据驱动故障诊断方法都是在监督学习的框架下设计的,这需要大量的标记样本。本文提出了一种新的自训练半监督深度学习(SSDL)方法,用于在少量标记样本和大量未标记样本的情况下训练故障诊断模型。寻址的SSDL方法是通过使用标记的样本初始化堆叠稀疏自编码器分类器来实现的,然后通过逐步从未标记的样本中采样一些具有最可靠伪标签的候选分类器来更新分类器。与现有自训练半监督框架中常用的静态抽样策略不同,在SSDL中提出了一种逐步开发机制,以逐步增加所选择的伪标记候选者的数量。此外,本文设计了一种基于距离的采样准则,代替预测精度作为伪标签的置信度估计,根据每个未标记样本在深度特征空间中的欧几里得距离,对其最近的标记样本分配标签。实验结果表明,与其他自训练半监督学习算法相比,所提出的SSDL方法具有较好的预测精度。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
A novel self-training semi-supervised deep learning approach for machinery fault diagnosis
Fault diagnosis is an indispensable basis for the collaborative maintenance in prognostic and health management. Most of existing data-driven fault diagnosis approaches are designed in the framework of supervised learning, which requires a large number of labelled samples. In this paper, a novel self-training semi-supervised deep learning (SSDL) approach is proposed to train a fault diagnosis model together with few labelled and abundant unlabelled samples. The addressed SSDL approach is realised by initialising a stacked sparse auto-encoder classifier using the labelled samples, and subsequently updating the classifier via sampling a few candidates with most reliable pseudo labels from the unlabelled samples step by step. Unlike the commonly used static sampling strategy in existing self-training semi-supervised frameworks, a gradually exploiting mechanism is proposed in SSDL to increase the number of selected pseudo-labelled candidates gradually. In addition, instead of using the prediction accuracy as the confidence estimation for pseudo-labels, a distance-based sampling criterion is designed to assign the label for each unlabelled sample by its nearest labelled sample based on their Euclidean distances in the deep feature space. The experimental results show that the proposed SSDL approach can achieve good prediction accuracy compared to other self-training semi-supervised learning algorithms.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
International Journal of Production Research
International Journal of Production Research 管理科学-工程:工业
CiteScore
19.20
自引率
14.10%
发文量
318
审稿时长
6.3 months
期刊介绍: The International Journal of Production Research (IJPR), published since 1961, is a well-established, highly successful and leading journal reporting manufacturing, production and operations management research. IJPR is published 24 times a year and includes papers on innovation management, design of products, manufacturing processes, production and logistics systems. Production economics, the essential behaviour of production resources and systems as well as the complex decision problems that arise in design, management and control of production and logistics systems are considered. IJPR is a journal for researchers and professors in mechanical engineering, industrial and systems engineering, operations research and management science, and business. It is also an informative reference for industrial managers looking to improve the efficiency and effectiveness of their production systems.
期刊最新文献
Order sequencing, tote scheduling, and robot routing optimization in multi-tote storage and retrieval autonomous mobile robot systems Scheduling stochastic distributed flexible job shops using an multi-objective evolutionary algorithm with simulation evaluation Customised product design optimisation considering module synergy effects and expert preferences A MINP model and hybrid heuristic algorithm for railway cold chain service network design problem: a case study of China Rescheduling optimisation of sustainable multi-objective fuzzy flexible job shop under uncertain environment
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1