Xuhui Huang, Obaid Elshafiey, Karim Farzia, L. Udpa, Ming Han, Y. Deng
{"title":"Acoustic Emission Source Localization using Deep Transfer Learning and Finite Element Modeling–based Knowledge Transfer","authors":"Xuhui Huang, Obaid Elshafiey, Karim Farzia, L. Udpa, Ming Han, Y. Deng","doi":"10.32548/2023.me-04348","DOIUrl":null,"url":null,"abstract":"This paper presents a novel data-driven approach to localize two types of acoustic emission sources in an aluminum plate, namely a Hsu-Nielsen source, which simulates a crack-like source, and steel ball impacts of varying diameters acting as the impact source. While deep neural networks have shown promise in previous studies, achieving high accuracy requires a large amount of training data, which may not always be feasible. To address this challenge, we investigated the applicability of transfer learning to address the issue of limited training data. Our approach involves transferring knowledge learned from numerical modeling to the experimental domain to localize nine different source locations. In the process, we evaluated six deep learning architectures using tenfold cross-validation and demonstrated the potential of transfer learning for efficient acoustic emission source localization, even with limited experimental data. This study contributes to the growing demand for running deep learning models with limited capacity and training time and highlights the promise of transfer learning methods such as fine-tuning pretrained models on large semi-related datasets.","PeriodicalId":49876,"journal":{"name":"Materials Evaluation","volume":null,"pages":null},"PeriodicalIF":0.5000,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Materials Evaluation","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.32548/2023.me-04348","RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"MATERIALS SCIENCE, CHARACTERIZATION & TESTING","Score":null,"Total":0}
引用次数: 0
Abstract
This paper presents a novel data-driven approach to localize two types of acoustic emission sources in an aluminum plate, namely a Hsu-Nielsen source, which simulates a crack-like source, and steel ball impacts of varying diameters acting as the impact source. While deep neural networks have shown promise in previous studies, achieving high accuracy requires a large amount of training data, which may not always be feasible. To address this challenge, we investigated the applicability of transfer learning to address the issue of limited training data. Our approach involves transferring knowledge learned from numerical modeling to the experimental domain to localize nine different source locations. In the process, we evaluated six deep learning architectures using tenfold cross-validation and demonstrated the potential of transfer learning for efficient acoustic emission source localization, even with limited experimental data. This study contributes to the growing demand for running deep learning models with limited capacity and training time and highlights the promise of transfer learning methods such as fine-tuning pretrained models on large semi-related datasets.
期刊介绍:
Materials Evaluation publishes articles, news and features intended to increase the NDT practitioner’s knowledge of the science and technology involved in the field, bringing informative articles to the NDT public while highlighting the ongoing efforts of ASNT to fulfill its mission. M.E. is a peer-reviewed journal, relying on technicians and researchers to help grow and educate its members by providing relevant, cutting-edge and exclusive content containing technical details and discussions. The only periodical of its kind, M.E. is circulated to members and nonmember paid subscribers. The magazine is truly international in scope, with readers in over 90 nations. The journal’s history and archive reaches back to the earliest formative days of the Society.