請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/80697完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 吳政忠(Tsung-Tsong Wu) | |
| dc.contributor.author | Hong-Yi Kang | en |
| dc.contributor.author | 康宏翊 | zh_TW |
| dc.date.accessioned | 2022-11-24T03:13:15Z | - |
| dc.date.available | 2021-11-05 | |
| dc.date.available | 2022-11-24T03:13:15Z | - |
| dc.date.copyright | 2021-11-05 | |
| dc.date.issued | 2021 | |
| dc.date.submitted | 2021-10-21 | |
| dc.identifier.citation | [1] T. Nehl and N. Demerdash, 'Application of finite element eddy current analysis to nondestructive detection of flaws in metallic structures,' IEEE Transactions on Magnetics, vol. 16, no. 5, pp. 1080-1082, 1980. [2] T.N. Grigsby and E.J. Tajchman, 'Properties of Lamb Waves Relevant to the Ultrasonic Inspection of Thin Plates,' IRE Transactions on Ultrasonic Engineering, vol. 8, no. 1, pp. 26-33, 1961. [3] R.B. Thompson, R.K. Elsley, W.E. Peterson, and C.F. Vasile, 'AN EMAT SYSTEM FOR DETECTING FLAWS IN STEAM GENERATOR TUBES,' IEEE Ultrasonics Symposium, Conference Paper, 1979. [4] J.M. Liu, 'Temperature Dependence of Elastic Stiffness in Aluminum Alloys Measured with Non-Contact Electromagnetic Acoustic Transducers (EMATS),' IEEE Ultrasonics Symposium, Conference Paper, 1984. [5] C. Edwards and S.B. Palmer, 'Wideband electromagnetic acoustic transducers utilising neodymium iron boron permanent magnets,' IEEE Transactions on Magnetics, vol. 26, no. 5, pp. 2080-2084, 1990. [6] A.V. Clark, S.R. Schaps and C.M. Fortunko, 'A well-shielded EMAT for on-line ultrasonic monitoring of GMA welding,' IEEE Ultrasonics Symposium, Conference Paper, 1991. [7] T. Yamasaki, S. Tamai and M. Hirao, 'Arrayed-coil EMAT for longitudinal wave in steel wires,' IEEE Ultrasonics Symposium. Proceedings (Cat. No. 98CH36102), Conference Paper, 1998. [8] H. Ogi, M. Hirao and T. Ohtani, 'Line-focusing electromagnetic acoustic transducers for the detection of slit defects,' IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 46, no. 2, pp. 341-346, 1999. [9] S. Legendre, D. Massicotte, J. Goyette and T.K. Bose, 'Wavelet-transform-based method of analysis for Lamb-wave ultrasonic NDE signals,' IEEE Transactions on Instrumentation and Measurement, vol. 49, no. 3, pp. 524-530, 2000. [10] P. Wilcox, M. Lowe and P. Cawley, 'Omnidirectional guided wave inspection of large metallic plate structures using an EMAT array,' IEEE Transactions on Instrumentation and Measurement, vol. 52, no. 4, pp. 653-665, 2005. [11] Sabato Marco Siniscalchi, Dong Yu, Li Deng and Chin-Hui Lee, 'Speech Recognition Using Long-Span Temporal Patterns in a Deep Network Model,' IEEE Signal Processing Letters, vol. 20, no. 3, pp. 201-204, 2013 [12] Martin Di Federico, Pedro Julián and Pablo S. Mandolesi, 'SCDVP: A Simplicial CNN Digital Visual Processor,' IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 61, no. 7, pp. 1962-1969, 2014 [13] Rodrigo Frassetto Nogueira, Roberto de Alencar Lotufo and Rubens Campos Machado, 'Fingerprint Liveness Detection Using Convolutional Neural Networks,' IEEE Transactions on Information Forensics and Security, vol. 11, no. 6, pp. 1206-1213, 2016. [14] Jiyong Chung and Keemin Sohn, 'Image-Based Learning to Measure Traffic Density Using a Deep Convolutional Neural Network,' IEEE Transactions on Intelligent Transportation Systems, vol. 19, no. 5, pp. 1670-1675, 2018. [15] S. Legendre, D. Massicotte, J. Goyette and T.K. Bose, 'Neural classification of Lamb wave ultrasonic weld testing signals using wavelet coefficients,' IEEE Transactions on Instrumentation and Measurement, vol. 50, no. 3, pp. 672-678, 2001. [16] Y. Yan, D. Liu, B. Gao, G. Y. Tian and Z. C. Cai, 'A Deep Learning-Based Ultrasonic Pattern Recognition Method for Inspecting Girth Weld Cracking of Gas Pipeline,' IEEE Sensors Journal, vol. 20, no. 14, 2020. [17] Hongyu Sun, Lisha Peng, Shen Wang, Songling Huang and Kaifeng Qu, 'Development of Frequency-Mixed Point-Focusing Shear Horizontal Guided-Wave EMAT for Defect Inspection Using Deep Neural Network,' IEEE Transactions on Instrumentation and Measurement, vol. 70, 2021. [18] Shujuan Wang, Lei Kang, Penghao Xin and Guofu Zhai, 'Characteristic research and analysis of EMAT's transduction efficiency for surface detection of aluminum plate,' IEEE International Conference on Electronic Measurement Instruments, Conference Paper, 2009. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/80697 | - |
| dc.description.abstract | 各式各樣的螺絲在生活周遭隨處可見,使用需求量非常大,但在生產螺絲的過程中,可發現有些螺絲有強度不足的現象發生,因為為待加工的螺絲基材本身就含有缺陷在裡面。因此,如何有效率地檢測螺絲基材中的缺陷,且避免有缺陷的材料加工成螺絲,對於螺絲加工廠都是一個重要的問題。而最普遍的解決方式為非破壞檢測,透過感測器激發出彈性波,進行波速的測量及試體缺陷的檢測,彈性波會在試體結構中傳遞,並再以感測器接收訊號,藉由此方法來判斷波速及缺陷訊號。 深度學習為人工智慧的分支,近年來廣泛的被運用在各領域,其中卷積類神經網路對於大量資料的特徵擷取有非常不錯的效果,在各種圖片及音訊的辨識都有優秀的表現,利用此人工智慧技術結合電磁超聲波檢測系統,探討類神經網路模型的運作模式,並設計出完善的資料前處理方法及模型最佳結構。 本論文所設計最佳的卷積類神經網路模型進行判斷螺絲基材缺陷的實驗,檢測螺絲基材直徑3.5 mm、5 mm及7 mm的缺陷,準確率分別為98.69%、98.71%及96.70%,同時,因為資料前處理所擷取的視窗範圍,可以確保輸入的訊號不會受到其他邊界回波的影響,模型因此可以得到較好的準確率,在實際應用上,本研究方法也非常貼近實際情況,訊號差別只在於後段邊界反射訊號的有無。 本研究架設了完整的實驗量測架構,藉由蒐集大量實驗訊號,設計了較適合模型的資料前處理方法以及最佳的模型結構,使檢測模型可以精準且快速的判別訊號,解決專業人員因為訊號的訊噪比太低難以判斷的困境,除此之外,利用兩種模型進行比較及調整,了解類神經網路模型的運作方式,搭配波動力學與相關知識,讓模型有較好的學習能力達到較高的準確率,讓此檢測方法可以廣泛地應用在螺絲基材的非破壞檢測,提升加工端的生產效率。 | zh_TW |
| dc.description.provenance | Made available in DSpace on 2022-11-24T03:13:15Z (GMT). No. of bitstreams: 1 U0001-1810202115073600.pdf: 2558355 bytes, checksum: d6a365aace42a4c55684daa8c8becfab (MD5) Previous issue date: 2021 | en |
| dc.description.tableofcontents | 致謝 I 中文摘要 II ABSTRACT IV 目錄 VI 表目錄 VIII 圖目錄 X 第一章 導論 1 1.1 研究動機 1 1.2 文獻回顧 2 1.3 章節介紹 4 第二章 雙電磁超聲波探頭檢測 7 2.1 實驗試體 7 2.2 電磁超聲波探頭 7 2.2.1 探頭特色及原理 7 2.2.2 探頭製作 8 2.2.3 探頭比較 9 2.3 雙電磁超聲波探頭量測實驗 10 2.3.1 量測架構 10 2.3.2 訊號量測 11 2.3.3 探頭距離 12 2.3.4 訊號頻率之探討 12 2.3.5 波速之量測 13 2.3.6 螺絲基材量測範圍 13 第三章 人工智慧模型 28 3.1 資料前處理 28 3.2 深度學習網路架構 29 3.2.1 深度類神經網路 29 3.2.2 卷積類神經網路 33 3.3 初始模型決定訊號前處理 34 3.3.1 深度類神經網路初始模型 34 3.3.2 卷積類神經網路初始模型 35 第四章 人工智慧結合雙電磁超聲波探頭 49 4.1 單一缺陷深度檢測模型 49 4.2 雙電磁超聲波探頭檢測模型 50 4.2.1 深度類神經網路模型 50 4.2.2 卷積類神經網路模型 52 4.3 模型比較 56 4.4 實際應用 57 第五章 結論與未來展望 74 5.1 結論 74 5.2 未來展望 76 參考文獻 77 | |
| dc.language.iso | zh-TW | |
| dc.subject | 卷積類神經網路 | zh_TW |
| dc.subject | 人工智慧 | zh_TW |
| dc.subject | 非破壞檢測 | zh_TW |
| dc.subject | 深度類神經網路 | zh_TW |
| dc.subject | 電磁超聲波 | zh_TW |
| dc.subject | Convolutional neural network | en |
| dc.subject | Artificial intelligence | en |
| dc.subject | Deep neural network | en |
| dc.subject | Nondestructive testing | en |
| dc.subject | Electromagnetic acoustic transducer | en |
| dc.title | 人工智慧結合雙電磁超聲波探頭檢測螺絲基材之缺陷 | zh_TW |
| dc.title | Integration of Artificial Intelligence and dual EMATs to detect defect in screw material | en |
| dc.date.schoolyear | 109-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.coadvisor | 劉佩玲(Pei-Ling Liu) | |
| dc.contributor.oralexamcommittee | 張瑞益(Hsin-Tsai Liu),孫嘉宏(Chih-Yang Tseng) | |
| dc.subject.keyword | 非破壞檢測,電磁超聲波,人工智慧,深度類神經網路,卷積類神經網路, | zh_TW |
| dc.subject.keyword | Nondestructive testing,Electromagnetic acoustic transducer,Artificial intelligence,Deep neural network,Convolutional neural network, | en |
| dc.relation.page | 79 | |
| dc.identifier.doi | 10.6342/NTU202103828 | |
| dc.rights.note | 同意授權(限校園內公開) | |
| dc.date.accepted | 2021-10-22 | |
| dc.contributor.author-college | 工學院 | zh_TW |
| dc.contributor.author-dept | 應用力學研究所 | zh_TW |
| 顯示於系所單位: | 應用力學研究所 | |
文件中的檔案:
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| U0001-1810202115073600.pdf 授權僅限NTU校內IP使用(校園外請利用VPN校外連線服務) | 2.5 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。
