In-process prediction of weld penetration depth using machine learning-based molten pool extraction technique in tungsten arc welding
Even though arc welding is widely utilized to join metallic parts with high reliability, the prediction and control of welding quality is challenging owing to difficulties in the prediction of weld penetration depth and the backside bead. In this study, an effective method for predicting weld penetr...
Gespeichert in:
Veröffentlicht in: | Journal of intelligent manufacturing 2024, Vol.35 (1), p.129-145 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 145 |
---|---|
container_issue | 1 |
container_start_page | 129 |
container_title | Journal of intelligent manufacturing |
container_volume | 35 |
creator | Baek, Daehyun Moon, Hyeong Soon Park, Sang-Hu |
description | Even though arc welding is widely utilized to join metallic parts with high reliability, the prediction and control of welding quality is challenging owing to difficulties in the prediction of weld penetration depth and the backside bead. In this study, an effective method for predicting weld penetration based on deep learning was proposed to control the welding quality in-process. The topside weld pool image was closely related to the welding quality and penetration depth and was also an accurate indicator of the state of welding over time. A prediction model for penetration depth using a topside weld pool image was constructed. Semantic segmentation based on a residual neural network was then performed on the acquired weld pool image. Consequently, an accurate weld pool shape was extracted. In addition, a penetration regression model was constructed based on a back-propagation neural network. Finally, the penetration depth (corresponding to the weld pool shape) was extracted via segmentation. The segmentation and regression models were combined to create a penetration prediction model. Considering a gas tungsten arc welding (GTAW) process, the predictions obtained from the proposed method were evaluated experimentally. In the validation process, the developed model quantitatively predicted the penetration depth in tungsten gas arc welding. The mean absolute error was 0.0596 mm with an
R
2
value of 0.9974. The model developed in this study can be utilized to predict weld depth penetration and in-processing time using surface images of the weld pool. |
doi_str_mv | 10.1007/s10845-022-02013-z |
format | Article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2914341671</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2914341671</sourcerecordid><originalsourceid>FETCH-LOGICAL-c319t-23b6dbc4e1e30b748648f8295d838931b58c627f2388bdff2cce86f13a26d20d3</originalsourceid><addsrcrecordid>eNp9kMtOwzAQRS0EEqXwA6wssTb4ESfOElU8KlViA2vLsSdtqtQJdiKge_4bt0Fix8IaeXTPnZmL0DWjt4zS4i4yqjJJKOfpUSbI_gTNmCw4USyTp2hGS5kTKZk8RxcxbimlpcrZDH0vPelDZyFG3AdwjR2azuOuxh_QOtyDhyGYY89BP2zwGBu_xjtjN40H3IIJPjVIZSI4vOvaATzuu67F8JnAyW0Au_HN-wi4SZ_Rr-NBZYI9Dkn4JTqrTRvh6rfO0dvjw-vimaxenpaL-xWxgpUD4aLKXWUzYCBoVWQqz1SteCmdEqoUrJLK5ryouVCqcnXNrQWV10wYnjtOnZijm8k3nZzWiYPedmPwaaTmJctExvKCJRWfVDZ0MQaodR-anQlfmlF9iFtPcesUtz7GrfcJEhMUk9ivIfxZ_0P9ALWwhkE</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2914341671</pqid></control><display><type>article</type><title>In-process prediction of weld penetration depth using machine learning-based molten pool extraction technique in tungsten arc welding</title><source>SpringerNature Journals</source><creator>Baek, Daehyun ; Moon, Hyeong Soon ; Park, Sang-Hu</creator><creatorcontrib>Baek, Daehyun ; Moon, Hyeong Soon ; Park, Sang-Hu</creatorcontrib><description>Even though arc welding is widely utilized to join metallic parts with high reliability, the prediction and control of welding quality is challenging owing to difficulties in the prediction of weld penetration depth and the backside bead. In this study, an effective method for predicting weld penetration based on deep learning was proposed to control the welding quality in-process. The topside weld pool image was closely related to the welding quality and penetration depth and was also an accurate indicator of the state of welding over time. A prediction model for penetration depth using a topside weld pool image was constructed. Semantic segmentation based on a residual neural network was then performed on the acquired weld pool image. Consequently, an accurate weld pool shape was extracted. In addition, a penetration regression model was constructed based on a back-propagation neural network. Finally, the penetration depth (corresponding to the weld pool shape) was extracted via segmentation. The segmentation and regression models were combined to create a penetration prediction model. Considering a gas tungsten arc welding (GTAW) process, the predictions obtained from the proposed method were evaluated experimentally. In the validation process, the developed model quantitatively predicted the penetration depth in tungsten gas arc welding. The mean absolute error was 0.0596 mm with an
R
2
value of 0.9974. The model developed in this study can be utilized to predict weld depth penetration and in-processing time using surface images of the weld pool.</description><identifier>ISSN: 0956-5515</identifier><identifier>EISSN: 1572-8145</identifier><identifier>DOI: 10.1007/s10845-022-02013-z</identifier><language>eng</language><publisher>New York: Springer US</publisher><subject>Arc welding machines ; Artificial neural networks ; Back propagation networks ; Business and Management ; Component reliability ; Control ; Deep learning ; Gas tungsten arc welding ; Image acquisition ; Image quality ; Image segmentation ; Machine learning ; Machines ; Manufacturing ; Mechatronics ; Melt pools ; Neural networks ; Penetration depth ; Prediction models ; Processes ; Production ; Regression models ; Robotics ; Semantic segmentation ; Tungsten ; Weld metal pool ; Welding</subject><ispartof>Journal of intelligent manufacturing, 2024, Vol.35 (1), p.129-145</ispartof><rights>The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c319t-23b6dbc4e1e30b748648f8295d838931b58c627f2388bdff2cce86f13a26d20d3</citedby><cites>FETCH-LOGICAL-c319t-23b6dbc4e1e30b748648f8295d838931b58c627f2388bdff2cce86f13a26d20d3</cites><orcidid>0000-0001-5332-1762</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s10845-022-02013-z$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s10845-022-02013-z$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,41488,42557,51319</link.rule.ids></links><search><creatorcontrib>Baek, Daehyun</creatorcontrib><creatorcontrib>Moon, Hyeong Soon</creatorcontrib><creatorcontrib>Park, Sang-Hu</creatorcontrib><title>In-process prediction of weld penetration depth using machine learning-based molten pool extraction technique in tungsten arc welding</title><title>Journal of intelligent manufacturing</title><addtitle>J Intell Manuf</addtitle><description>Even though arc welding is widely utilized to join metallic parts with high reliability, the prediction and control of welding quality is challenging owing to difficulties in the prediction of weld penetration depth and the backside bead. In this study, an effective method for predicting weld penetration based on deep learning was proposed to control the welding quality in-process. The topside weld pool image was closely related to the welding quality and penetration depth and was also an accurate indicator of the state of welding over time. A prediction model for penetration depth using a topside weld pool image was constructed. Semantic segmentation based on a residual neural network was then performed on the acquired weld pool image. Consequently, an accurate weld pool shape was extracted. In addition, a penetration regression model was constructed based on a back-propagation neural network. Finally, the penetration depth (corresponding to the weld pool shape) was extracted via segmentation. The segmentation and regression models were combined to create a penetration prediction model. Considering a gas tungsten arc welding (GTAW) process, the predictions obtained from the proposed method were evaluated experimentally. In the validation process, the developed model quantitatively predicted the penetration depth in tungsten gas arc welding. The mean absolute error was 0.0596 mm with an
R
2
value of 0.9974. The model developed in this study can be utilized to predict weld depth penetration and in-processing time using surface images of the weld pool.</description><subject>Arc welding machines</subject><subject>Artificial neural networks</subject><subject>Back propagation networks</subject><subject>Business and Management</subject><subject>Component reliability</subject><subject>Control</subject><subject>Deep learning</subject><subject>Gas tungsten arc welding</subject><subject>Image acquisition</subject><subject>Image quality</subject><subject>Image segmentation</subject><subject>Machine learning</subject><subject>Machines</subject><subject>Manufacturing</subject><subject>Mechatronics</subject><subject>Melt pools</subject><subject>Neural networks</subject><subject>Penetration depth</subject><subject>Prediction models</subject><subject>Processes</subject><subject>Production</subject><subject>Regression models</subject><subject>Robotics</subject><subject>Semantic segmentation</subject><subject>Tungsten</subject><subject>Weld metal pool</subject><subject>Welding</subject><issn>0956-5515</issn><issn>1572-8145</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNp9kMtOwzAQRS0EEqXwA6wssTb4ESfOElU8KlViA2vLsSdtqtQJdiKge_4bt0Fix8IaeXTPnZmL0DWjt4zS4i4yqjJJKOfpUSbI_gTNmCw4USyTp2hGS5kTKZk8RxcxbimlpcrZDH0vPelDZyFG3AdwjR2azuOuxh_QOtyDhyGYY89BP2zwGBu_xjtjN40H3IIJPjVIZSI4vOvaATzuu67F8JnAyW0Au_HN-wi4SZ_Rr-NBZYI9Dkn4JTqrTRvh6rfO0dvjw-vimaxenpaL-xWxgpUD4aLKXWUzYCBoVWQqz1SteCmdEqoUrJLK5ryouVCqcnXNrQWV10wYnjtOnZijm8k3nZzWiYPedmPwaaTmJctExvKCJRWfVDZ0MQaodR-anQlfmlF9iFtPcesUtz7GrfcJEhMUk9ivIfxZ_0P9ALWwhkE</recordid><startdate>2024</startdate><enddate>2024</enddate><creator>Baek, Daehyun</creator><creator>Moon, Hyeong Soon</creator><creator>Park, Sang-Hu</creator><general>Springer US</general><general>Springer Nature B.V</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7TB</scope><scope>8FD</scope><scope>FR3</scope><scope>JQ2</scope><scope>K9.</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><orcidid>https://orcid.org/0000-0001-5332-1762</orcidid></search><sort><creationdate>2024</creationdate><title>In-process prediction of weld penetration depth using machine learning-based molten pool extraction technique in tungsten arc welding</title><author>Baek, Daehyun ; Moon, Hyeong Soon ; Park, Sang-Hu</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c319t-23b6dbc4e1e30b748648f8295d838931b58c627f2388bdff2cce86f13a26d20d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Arc welding machines</topic><topic>Artificial neural networks</topic><topic>Back propagation networks</topic><topic>Business and Management</topic><topic>Component reliability</topic><topic>Control</topic><topic>Deep learning</topic><topic>Gas tungsten arc welding</topic><topic>Image acquisition</topic><topic>Image quality</topic><topic>Image segmentation</topic><topic>Machine learning</topic><topic>Machines</topic><topic>Manufacturing</topic><topic>Mechatronics</topic><topic>Melt pools</topic><topic>Neural networks</topic><topic>Penetration depth</topic><topic>Prediction models</topic><topic>Processes</topic><topic>Production</topic><topic>Regression models</topic><topic>Robotics</topic><topic>Semantic segmentation</topic><topic>Tungsten</topic><topic>Weld metal pool</topic><topic>Welding</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Baek, Daehyun</creatorcontrib><creatorcontrib>Moon, Hyeong Soon</creatorcontrib><creatorcontrib>Park, Sang-Hu</creatorcontrib><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Mechanical & Transportation Engineering Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>Journal of intelligent manufacturing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Baek, Daehyun</au><au>Moon, Hyeong Soon</au><au>Park, Sang-Hu</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>In-process prediction of weld penetration depth using machine learning-based molten pool extraction technique in tungsten arc welding</atitle><jtitle>Journal of intelligent manufacturing</jtitle><stitle>J Intell Manuf</stitle><date>2024</date><risdate>2024</risdate><volume>35</volume><issue>1</issue><spage>129</spage><epage>145</epage><pages>129-145</pages><issn>0956-5515</issn><eissn>1572-8145</eissn><abstract>Even though arc welding is widely utilized to join metallic parts with high reliability, the prediction and control of welding quality is challenging owing to difficulties in the prediction of weld penetration depth and the backside bead. In this study, an effective method for predicting weld penetration based on deep learning was proposed to control the welding quality in-process. The topside weld pool image was closely related to the welding quality and penetration depth and was also an accurate indicator of the state of welding over time. A prediction model for penetration depth using a topside weld pool image was constructed. Semantic segmentation based on a residual neural network was then performed on the acquired weld pool image. Consequently, an accurate weld pool shape was extracted. In addition, a penetration regression model was constructed based on a back-propagation neural network. Finally, the penetration depth (corresponding to the weld pool shape) was extracted via segmentation. The segmentation and regression models were combined to create a penetration prediction model. Considering a gas tungsten arc welding (GTAW) process, the predictions obtained from the proposed method were evaluated experimentally. In the validation process, the developed model quantitatively predicted the penetration depth in tungsten gas arc welding. The mean absolute error was 0.0596 mm with an
R
2
value of 0.9974. The model developed in this study can be utilized to predict weld depth penetration and in-processing time using surface images of the weld pool.</abstract><cop>New York</cop><pub>Springer US</pub><doi>10.1007/s10845-022-02013-z</doi><tpages>17</tpages><orcidid>https://orcid.org/0000-0001-5332-1762</orcidid></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0956-5515 |
ispartof | Journal of intelligent manufacturing, 2024, Vol.35 (1), p.129-145 |
issn | 0956-5515 1572-8145 |
language | eng |
recordid | cdi_proquest_journals_2914341671 |
source | SpringerNature Journals |
subjects | Arc welding machines Artificial neural networks Back propagation networks Business and Management Component reliability Control Deep learning Gas tungsten arc welding Image acquisition Image quality Image segmentation Machine learning Machines Manufacturing Mechatronics Melt pools Neural networks Penetration depth Prediction models Processes Production Regression models Robotics Semantic segmentation Tungsten Weld metal pool Welding |
title | In-process prediction of weld penetration depth using machine learning-based molten pool extraction technique in tungsten arc welding |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T06%3A25%3A41IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=In-process%20prediction%20of%20weld%20penetration%20depth%20using%20machine%20learning-based%20molten%20pool%20extraction%20technique%20in%20tungsten%20arc%20welding&rft.jtitle=Journal%20of%20intelligent%20manufacturing&rft.au=Baek,%20Daehyun&rft.date=2024&rft.volume=35&rft.issue=1&rft.spage=129&rft.epage=145&rft.pages=129-145&rft.issn=0956-5515&rft.eissn=1572-8145&rft_id=info:doi/10.1007/s10845-022-02013-z&rft_dat=%3Cproquest_cross%3E2914341671%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2914341671&rft_id=info:pmid/&rfr_iscdi=true |