基于马尔可夫转换场和变压器网络的电缆绝缘缺陷识别方法研究

IF 1.9 3区 物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY
Ning Zhao, Yongyi Fang, Siying Wang, Qian Li, Xiaonan Wang, Chi Feng
{"title":"基于马尔可夫转换场和变压器网络的电缆绝缘缺陷识别方法研究","authors":"Ning Zhao, Yongyi Fang, Siying Wang, Qian Li, Xiaonan Wang, Chi Feng","doi":"10.3389/fphy.2024.1432783","DOIUrl":null,"url":null,"abstract":"Identifying cable insulation defects is crucial for preventing system failures and ensuring the reliability of electrical infrastructure. This paper introduces a novel method leveraging the Markov transition field (MTF) and Transformer network to improve the precision of cable insulation defect identification and enhance the algorithm's noise resistance. Firstly, the algorithm performs modal transformation on the time series data acquired by the ultrasonic probe through MTF, generating corresponding images. Following this, the image data are input into a pre-trained Transformer network to achieve automated feature extraction. Subsequently, a multi-head attention mechanism is introduced, which assigns weights to the features extracted by the Transformer network, thereby emphasizing the most critical information for the identification task. Finally, more accurate defect identification is achieved based on the weighted features. The results demonstrate that this method achieves higher accuracy and stronger noise resistance compared to traditional image processing and recognition methods, making it a robust solution for cable insulation defect identification.","PeriodicalId":12507,"journal":{"name":"Frontiers in Physics","volume":null,"pages":null},"PeriodicalIF":1.9000,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Research on the identification method of cable insulation defects based on Markov transition fields and transformer networks\",\"authors\":\"Ning Zhao, Yongyi Fang, Siying Wang, Qian Li, Xiaonan Wang, Chi Feng\",\"doi\":\"10.3389/fphy.2024.1432783\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Identifying cable insulation defects is crucial for preventing system failures and ensuring the reliability of electrical infrastructure. This paper introduces a novel method leveraging the Markov transition field (MTF) and Transformer network to improve the precision of cable insulation defect identification and enhance the algorithm's noise resistance. Firstly, the algorithm performs modal transformation on the time series data acquired by the ultrasonic probe through MTF, generating corresponding images. Following this, the image data are input into a pre-trained Transformer network to achieve automated feature extraction. Subsequently, a multi-head attention mechanism is introduced, which assigns weights to the features extracted by the Transformer network, thereby emphasizing the most critical information for the identification task. Finally, more accurate defect identification is achieved based on the weighted features. The results demonstrate that this method achieves higher accuracy and stronger noise resistance compared to traditional image processing and recognition methods, making it a robust solution for cable insulation defect identification.\",\"PeriodicalId\":12507,\"journal\":{\"name\":\"Frontiers in Physics\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.9000,\"publicationDate\":\"2024-08-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers in Physics\",\"FirstCategoryId\":\"101\",\"ListUrlMain\":\"https://doi.org/10.3389/fphy.2024.1432783\",\"RegionNum\":3,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"PHYSICS, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Physics","FirstCategoryId":"101","ListUrlMain":"https://doi.org/10.3389/fphy.2024.1432783","RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PHYSICS, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

摘要

识别电缆绝缘缺陷对于防止系统故障和确保电力基础设施的可靠性至关重要。本文介绍了一种利用马尔可夫变换场(MTF)和变压器网络的新方法,以提高电缆绝缘缺陷识别的精度并增强算法的抗噪声能力。首先,该算法通过 MTF 对超声波探头获取的时间序列数据进行模态变换,生成相应的图像。然后,将图像数据输入预先训练好的 Transformer 网络,实现自动特征提取。随后,引入多头关注机制,为 Transformer 网络提取的特征分配权重,从而强调识别任务中最关键的信息。最后,根据加权特征实现了更准确的缺陷识别。研究结果表明,与传统的图像处理和识别方法相比,该方法具有更高的准确性和更强的抗噪能力,是电缆绝缘缺陷识别的可靠解决方案。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Research on the identification method of cable insulation defects based on Markov transition fields and transformer networks
Identifying cable insulation defects is crucial for preventing system failures and ensuring the reliability of electrical infrastructure. This paper introduces a novel method leveraging the Markov transition field (MTF) and Transformer network to improve the precision of cable insulation defect identification and enhance the algorithm's noise resistance. Firstly, the algorithm performs modal transformation on the time series data acquired by the ultrasonic probe through MTF, generating corresponding images. Following this, the image data are input into a pre-trained Transformer network to achieve automated feature extraction. Subsequently, a multi-head attention mechanism is introduced, which assigns weights to the features extracted by the Transformer network, thereby emphasizing the most critical information for the identification task. Finally, more accurate defect identification is achieved based on the weighted features. The results demonstrate that this method achieves higher accuracy and stronger noise resistance compared to traditional image processing and recognition methods, making it a robust solution for cable insulation defect identification.
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
Frontiers in Physics
Frontiers in Physics Mathematics-Mathematical Physics
CiteScore
4.50
自引率
6.50%
发文量
1215
审稿时长
12 weeks
期刊介绍: Frontiers in Physics publishes rigorously peer-reviewed research across the entire field, from experimental, to computational and theoretical physics. This multidisciplinary open-access journal is at the forefront of disseminating and communicating scientific knowledge and impactful discoveries to researchers, academics, engineers and the public worldwide.
×
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