An unsupervised machine learning algorithm for in-situ defect-detection in laser powder-bed fusion
This research studies the development of a machine-learning algorithm to detect porosity induced during laser powder bed fusion (LPBF) due to the lack of fusion (LoF) phenomenon. The detection algorithm is based on analyzing in-situ light intensity data emitted from the melt pool during LPBF. A Self...
Uložené v:
| Vydané v: | Journal of manufacturing processes Ročník 81; s. 476 - 489 |
|---|---|
| Hlavní autori: | , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Elsevier Ltd
01.09.2022
|
| Predmet: | |
| ISSN: | 1526-6125, 2212-4616 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | This research studies the development of a machine-learning algorithm to detect porosity induced during laser powder bed fusion (LPBF) due to the lack of fusion (LoF) phenomenon. The detection algorithm is based on analyzing in-situ light intensity data emitted from the melt pool during LPBF. A Self-Organizing Map (SOM), an unsupervised machine learning algorithm, is thoroughly customized to classify disturbances in the light intensity signal, where the clustered disturbances are mapped with the geometrical feature and position of defects. The defects' position and size are correlated with the actual geometrical feature and position of defects identified through a post-processing computed tomography (CT) scanning. To this end, two sets of samples are designed, including 1) samples with intentional micro-voids to mimic the lack of fusion in the printed parts for assessing the sensor response and calibrating/optimizing the SOM algorithm, and 2) samples with randomized pores induced normally during the process to evaluate the proposed SOM algorithm at different process parameters. Along with the SOM approach, a volumetric segmentation method and confusion matrix are incorporated into the SOM-based algorithm to examine the sensitivity (true positive rates) and specificity (true negative rates) of defect prediction. The comparison between the outcome of the prediction algorithm and experimental CT data indicates that the sensitivity and specificity are from 61 % to 94 % and 69 % to 93 %, respectively, where the prediction percentage is highly dependent on the process parameters. In comparison with conventional prediction algorithms (e.g., Absolute Limits), the proposed SOM algorithm has a higher prediction rate when it is faster and can rapidly identify the location of defects that may open up an opportunity to develop intermittent controllers for LPBF. |
|---|---|
| AbstractList | This research studies the development of a machine-learning algorithm to detect porosity induced during laser powder bed fusion (LPBF) due to the lack of fusion (LoF) phenomenon. The detection algorithm is based on analyzing in-situ light intensity data emitted from the melt pool during LPBF. A Self-Organizing Map (SOM), an unsupervised machine learning algorithm, is thoroughly customized to classify disturbances in the light intensity signal, where the clustered disturbances are mapped with the geometrical feature and position of defects. The defects' position and size are correlated with the actual geometrical feature and position of defects identified through a post-processing computed tomography (CT) scanning. To this end, two sets of samples are designed, including 1) samples with intentional micro-voids to mimic the lack of fusion in the printed parts for assessing the sensor response and calibrating/optimizing the SOM algorithm, and 2) samples with randomized pores induced normally during the process to evaluate the proposed SOM algorithm at different process parameters. Along with the SOM approach, a volumetric segmentation method and confusion matrix are incorporated into the SOM-based algorithm to examine the sensitivity (true positive rates) and specificity (true negative rates) of defect prediction. The comparison between the outcome of the prediction algorithm and experimental CT data indicates that the sensitivity and specificity are from 61 % to 94 % and 69 % to 93 %, respectively, where the prediction percentage is highly dependent on the process parameters. In comparison with conventional prediction algorithms (e.g., Absolute Limits), the proposed SOM algorithm has a higher prediction rate when it is faster and can rapidly identify the location of defects that may open up an opportunity to develop intermittent controllers for LPBF. |
| Author | Eischer, Christopher Taherkhani, Katayoon Toyserkani, Ehsan |
| Author_xml | – sequence: 1 givenname: Katayoon surname: Taherkhani fullname: Taherkhani, Katayoon organization: Multi-scale Additive Manufacturing Lab, Department of Mechanical and Mechatronics Engineering, University of Waterloo, Canada – sequence: 2 givenname: Christopher surname: Eischer fullname: Eischer, Christopher organization: Electro-Optical Systems (EOS), Krailling, Germany – sequence: 3 givenname: Ehsan surname: Toyserkani fullname: Toyserkani, Ehsan email: ehsan.toyserkani@uwaterloo.ca organization: Multi-scale Additive Manufacturing Lab, Department of Mechanical and Mechatronics Engineering, University of Waterloo, Canada |
| BookMark | eNqFkL1OwzAUhS1UJFrgDRj8AgnXduoEBqSq4k-qxAKz5djXraPEqeykiLcnVZkYYDrD0Xd077cgs9AHJOSGQc6Aydsmbzq9j33OgfMcZA5lcUbmnDOeFZLJGZmzJZeZZHx5QRYpNQCMF8DmpF4FOoY07jEefEJLO212PiBtUcfgw5bqdttHP-w66vpIfciSH0Zq0aEZMovDFL4PU0FbnTDSff9pMWb1tOXGNFVX5NzpNuH1T16Sj6fH9_VLtnl7fl2vNpkRIIes4qiFLSWAAItoamGcMRWzABW3jGl-x0AbJnRZ1YY5o4VDrHRdFw6EKMUluT_tmtinFNEp4wd9PG6I2reKgTraUo062VJHWwqkmmxNcPEL3kff6fj1H_ZwwnB67OAxqmQ8BoPWx0mMsr3_e-Ab_f2LiQ |
| CitedBy_id | crossref_primary_10_1016_j_mfglet_2025_06_117 crossref_primary_10_1177_09544089251345640 crossref_primary_10_1016_j_measurement_2025_117413 crossref_primary_10_1016_j_optlaseng_2025_109194 crossref_primary_10_1007_s10845_024_02490_4 crossref_primary_10_1002_eem2_70041 crossref_primary_10_1007_s10845_023_02232_y crossref_primary_10_1007_s00170_025_15273_9 crossref_primary_10_3390_inventions9040087 crossref_primary_10_1016_j_jmapro_2024_02_006 crossref_primary_10_1007_s00170_025_16164_9 crossref_primary_10_1016_j_jmapro_2023_12_024 crossref_primary_10_1007_s00170_025_16431_9 crossref_primary_10_1016_j_jmapro_2025_03_107 crossref_primary_10_1016_j_optlaseng_2024_108522 crossref_primary_10_1088_2631_7990_ad65cd crossref_primary_10_3390_jmmp8050197 crossref_primary_10_1016_j_addma_2024_104354 crossref_primary_10_1016_j_measurement_2024_114877 crossref_primary_10_1016_j_jmapro_2023_05_030 crossref_primary_10_1007_s00170_023_12663_9 crossref_primary_10_1007_s10311_024_01741_3 crossref_primary_10_1016_j_mtcomm_2022_104900 crossref_primary_10_1007_s41939_024_00658_2 crossref_primary_10_1007_s00170_025_15777_4 crossref_primary_10_1371_journal_pone_0316996 crossref_primary_10_1142_S201032472540003X crossref_primary_10_1016_j_engappai_2024_109058 crossref_primary_10_1007_s00170_025_15507_w crossref_primary_10_1016_j_addma_2023_103894 crossref_primary_10_1016_j_matdes_2024_112850 crossref_primary_10_1016_j_heliyon_2024_e35047 crossref_primary_10_1016_j_jmapro_2023_05_048 crossref_primary_10_3390_ma17133287 crossref_primary_10_1007_s10845_023_02170_9 crossref_primary_10_1016_j_ijmecsci_2023_108102 crossref_primary_10_1016_j_mfglet_2025_06_156 |
| Cites_doi | 10.1007/s00170-018-1728-0 10.1016/j.aej.2014.09.007 10.1080/24725854.2017.1417656 10.1016/j.matdes.2018.07.002 10.1016/j.ijleo.2016.01.070 10.1016/j.jmatprotec.2015.12.024 10.1109/ACCESS.2021.3067302 10.1016/j.measurement.2018.12.067 10.1117/1.OE.54.1.011008 10.1007/s40964-019-00108-3 10.1016/j.cirp.2019.03.021 10.1016/j.jmatprotec.2011.09.020 10.4028/www.scientific.net/KEM.554-557.1828 10.1016/j.phpro.2010.08.078 10.1007/s10845-018-1451-6 10.1016/j.ijfatigue.2019.105194 10.1016/j.jmsy.2019.02.005 10.1016/j.actamat.2016.05.017 10.4028/www.scientific.net/MSF.834.93 10.1080/00207543.2014.916431 10.1016/j.asoc.2007.02.002 10.1186/s12864-019-6413-7 10.1016/j.rcim.2017.07.001 10.1016/j.matdes.2016.01.099 10.1016/j.jmatprotec.2009.07.017 10.1016/j.phpro.2011.03.085 10.1007/s11665-018-3690-2 10.1080/00224065.2018.1507563 10.1115/1.4028540 10.1115/1.4038598 |
| ContentType | Journal Article |
| Copyright | 2022 The Society of Manufacturing Engineers |
| Copyright_xml | – notice: 2022 The Society of Manufacturing Engineers |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.jmapro.2022.06.074 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 2212-4616 |
| EndPage | 489 |
| ExternalDocumentID | 10_1016_j_jmapro_2022_06_074 S1526612522004534 |
| GroupedDBID | --K --M .~1 0R~ 1B1 1~. 1~5 29K 3V. 4.4 457 4G. 5GY 5VS 7-5 71M 7WY 883 88I 8AO 8FE 8FG 8FL 8FW 8P~ 8R4 8R5 9M8 AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO ABFNM ABJCF ABJNI ABMAC ABUWG ABXDB ABYKQ ACDAQ ACGFS ACGOD ACIWK ACNNM ACRLP ADBBV ADEZE ADMUD ADTZH AEBSH AECPX AEKER AENEX AFFNX AFKRA AFKWA AFTJW AGHFR AGUBO AGYEJ AHJVU AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ ASPBG AVWKF AXJTR AZFZN AZQEC BENPR BEZIV BGLVJ BJAXD BKOJK BLXMC BPHCQ CCPQU CS3 D-I DU5 DWQXO E3Z EBS EFJIC EFLBG EJD EP2 EP3 FDB FEDTE FGOYB FIRID FNPLU FRNLG FYGXN GBLVA GNUQQ GROUPED_ABI_INFORM_COMPLETE GROUPED_ABI_INFORM_RESEARCH HCIFZ HVGLF HZ~ H~9 J1W JJJVA K60 K6~ KOM L6V M0C M0F M2P M41 M7S MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. PQBIZ PQQKQ PROAC PTHSS Q2X Q38 R2- RIG RNS ROL RWL S0X SDF SES SPC SPCBC SST SSZ T5K TAE TN5 U5U ~G- 9DU AATTM AAXKI AAYWO AAYXX ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFFHD AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS PHGZM PHGZT PQBZA PQGLB ~HD |
| ID | FETCH-LOGICAL-c306t-82ea3d760030deecb3cfcc81d0082d11a2910ac13a78bc1fca3fee8abb4f03373 |
| ISICitedReferencesCount | 47 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000878519800002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1526-6125 |
| IngestDate | Sat Nov 29 07:05:10 EST 2025 Tue Nov 18 22:31:06 EST 2025 Fri Feb 23 02:38:09 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Laser powder bed fusion Quality assurance Additive manufacturing Clustering algorithm In-situ sensors Machine learning |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c306t-82ea3d760030deecb3cfcc81d0082d11a2910ac13a78bc1fca3fee8abb4f03373 |
| PageCount | 14 |
| ParticipantIDs | crossref_citationtrail_10_1016_j_jmapro_2022_06_074 crossref_primary_10_1016_j_jmapro_2022_06_074 elsevier_sciencedirect_doi_10_1016_j_jmapro_2022_06_074 |
| PublicationCentury | 2000 |
| PublicationDate | September 2022 2022-09-00 |
| PublicationDateYYYYMMDD | 2022-09-01 |
| PublicationDate_xml | – month: 09 year: 2022 text: September 2022 |
| PublicationDecade | 2020 |
| PublicationTitle | Journal of manufacturing processes |
| PublicationYear | 2022 |
| Publisher | Elsevier Ltd |
| Publisher_xml | – name: Elsevier Ltd |
| References | Taherkhani, Sheydaeian, Eischer, Otto, Toyserkani (bb0220) Oct. 2021; 46 Gestel (bb0025) 2015 Gobert, Reutzel, Petrich, Nassar, Phoha (bb0055) 2018; 21 Bayle, Doubenskaia (bb0020) 2008; 698505 Everton (bb0015) Apr. 2016; 95 Grasso, Colosimo, Pacella (bb0090) 2014; 52 Kwon (bb0210) 2020; 31 Zhang (bb0065) 2019; 128 Shevchik, Kenel, Leinenbach, Wasmer (bb0110) 2018; 21 Craeghs, Bechmann, Berumen, Kruth (bb0050) 2010; 5 Knaak, Masseling, Duong, Abels, Gillner (bb0115) 2021; 9 Toyserkani, Sarker, Ibhadode, Liravi, Russo, Taherkhani (bb0005) 2021 Chaudhary, Bhatia, Ahlawat (bb0240) 2014 Jafari-Marandi, Khanzadeh, Tian, Smith, Bian (bb0175) 2019; 51 Furumoto, Ueda, Kobayashi, Yassin, Hosokawa, Abe (bb0160) 2009; 209 Wu, Yu, Wang (bb0185) 2019; 136 Caggiano, Zhang, Alfieri, Caiazzo, Gao, Teti (bb0150) 2019; 68 Fuchs, Eischer (bb0225) 2018 Chicco, Jurman (bb0245) 2020; 21 Ferrar, Mullen, Jones, Stamp, Sutcliffe (bb0255) 2012; 212 Jacobsmuhlen, Kleszczynski, Schneider, Witt (bb0145) 2013 Khanzadeh, Chowdhury, Tschopp, Doude, Marufuzzaman, Bian (bb0170) 2019; 51 Kamimura (bb0235) 2011 Scime, Beuth (bb0070) 2018; 19 Wasmer, Kenel, Leinenbach, Shevchik (bb0195) 2018 Gaikwad, Imani, Rao, Yang, Reutzel (bb0155) 2019; 20 Petrich, Gobert, Phoha, Nassar, Reutzel (bb0125) 2017 Kanko, Sibley, Fraser (bb0215) 2016; 231 Okaro, Jayasinghe, Sutcliffe, Black, Paoletti, Green (bb0100) 2019; 27 Toyserkani, Khajepour, Corbin (bb0140) 2004 Haykin (bb0230) 1999 Matthews, Guss, Khairallah, Rubenchik, Depond, King (bb0200) 2016; 114 Lott, Schleifenbaum, Meiners, Wissenbach, Hinke, Bültmann (bb0075) 2011; 12 Abdelrahman, Reutzel, Nassar, Starr (bb0040) 2017; 15 Van Belle, Vansteenkiste, Boyer (bb0035) 2013; 554 Scime, Beuth (bb0060) 2019; 25 Gong, Rafi, Starr, Stucker (bb0205) 2013 Ye (bb0030) 2018; 96 Thombansen, Gatej, Pereira (bb0080) 2014; 54 Alimardani, Toyserkani (bb0130) 2008; 8 Baumgartl, Tomas, Buettner, Merkel (bb0165) 2020; 5 Khanzadeh, Rao, Jafari-Marandi, Smith, Tschopp, Bian (bb0180) 2018; 140 Zhang, Hong, Ye, Zhu, Fuh (bb0105) 2018; 156 Reijonen, Revuelta, Riipinen, Ruusuvuori, Puukko (bb0250) 2020; 32 Doubenskaia, Smurov, Teleshevskiy, Bertrand, Zhirnov (bb0085) 2015; 834 Grasso, Demir, Previtali, Colosimo (bb0045) 2018; 49 Tapia, Elwany (bb0010) 2014; 136 Sutton, Barto (bb0190) 2018; 2018 Wasmer, Le-Quang, Meylan, Shevchik (bb0120) Feb. 2019; 28 Sohrabpoor (bb0135) 2016; 127 Colosimo, Grasso (bb0095) 2018; 50 Zhang (10.1016/j.jmapro.2022.06.074_bb0065) 2019; 128 Wasmer (10.1016/j.jmapro.2022.06.074_bb0195) 2018 Shevchik (10.1016/j.jmapro.2022.06.074_bb0110) 2018; 21 Jacobsmuhlen (10.1016/j.jmapro.2022.06.074_bb0145) 2013 Kanko (10.1016/j.jmapro.2022.06.074_bb0215) 2016; 231 Gestel (10.1016/j.jmapro.2022.06.074_bb0025) 2015 Kwon (10.1016/j.jmapro.2022.06.074_bb0210) 2020; 31 Everton (10.1016/j.jmapro.2022.06.074_bb0015) 2016; 95 Scime (10.1016/j.jmapro.2022.06.074_bb0070) 2018; 19 Bayle (10.1016/j.jmapro.2022.06.074_bb0020) 2008; 698505 Grasso (10.1016/j.jmapro.2022.06.074_bb0090) 2014; 52 Gong (10.1016/j.jmapro.2022.06.074_bb0205) 2013 Haykin (10.1016/j.jmapro.2022.06.074_bb0230) 1999 Colosimo (10.1016/j.jmapro.2022.06.074_bb0095) 2018; 50 Baumgartl (10.1016/j.jmapro.2022.06.074_bb0165) 2020; 5 Reijonen (10.1016/j.jmapro.2022.06.074_bb0250) 2020; 32 Okaro (10.1016/j.jmapro.2022.06.074_bb0100) 2019; 27 Toyserkani (10.1016/j.jmapro.2022.06.074_bb0005) 2021 Khanzadeh (10.1016/j.jmapro.2022.06.074_bb0180) 2018; 140 Doubenskaia (10.1016/j.jmapro.2022.06.074_bb0085) 2015; 834 Alimardani (10.1016/j.jmapro.2022.06.074_bb0130) 2008; 8 Knaak (10.1016/j.jmapro.2022.06.074_bb0115) 2021; 9 Sutton (10.1016/j.jmapro.2022.06.074_bb0190) 2018; 2018 Caggiano (10.1016/j.jmapro.2022.06.074_bb0150) 2019; 68 Craeghs (10.1016/j.jmapro.2022.06.074_bb0050) 2010; 5 Wasmer (10.1016/j.jmapro.2022.06.074_bb0120) 2019; 28 Khanzadeh (10.1016/j.jmapro.2022.06.074_bb0170) 2019; 51 Lott (10.1016/j.jmapro.2022.06.074_bb0075) 2011; 12 Tapia (10.1016/j.jmapro.2022.06.074_bb0010) 2014; 136 Taherkhani (10.1016/j.jmapro.2022.06.074_bb0220) 2021; 46 Gobert (10.1016/j.jmapro.2022.06.074_bb0055) 2018; 21 Jafari-Marandi (10.1016/j.jmapro.2022.06.074_bb0175) 2019; 51 Zhang (10.1016/j.jmapro.2022.06.074_bb0105) 2018; 156 Ferrar (10.1016/j.jmapro.2022.06.074_bb0255) 2012; 212 Abdelrahman (10.1016/j.jmapro.2022.06.074_bb0040) 2017; 15 Grasso (10.1016/j.jmapro.2022.06.074_bb0045) 2018; 49 Fuchs (10.1016/j.jmapro.2022.06.074_bb0225) 2018 Petrich (10.1016/j.jmapro.2022.06.074_bb0125) 2017 Kamimura (10.1016/j.jmapro.2022.06.074_bb0235) 2011 Chicco (10.1016/j.jmapro.2022.06.074_bb0245) 2020; 21 Van Belle (10.1016/j.jmapro.2022.06.074_bb0035) 2013; 554 Sohrabpoor (10.1016/j.jmapro.2022.06.074_bb0135) 2016; 127 Chaudhary (10.1016/j.jmapro.2022.06.074_bb0240) 2014 Wu (10.1016/j.jmapro.2022.06.074_bb0185) 2019; 136 Scime (10.1016/j.jmapro.2022.06.074_bb0060) 2019; 25 Toyserkani (10.1016/j.jmapro.2022.06.074_bb0140) 2004 Gaikwad (10.1016/j.jmapro.2022.06.074_bb0155) 2019; 20 Ye (10.1016/j.jmapro.2022.06.074_bb0030) 2018; 96 Matthews (10.1016/j.jmapro.2022.06.074_bb0200) 2016; 114 Thombansen (10.1016/j.jmapro.2022.06.074_bb0080) 2014; 54 Furumoto (10.1016/j.jmapro.2022.06.074_bb0160) 2009; 209 |
| References_xml | – volume: 140 year: 2018 ident: bb0180 article-title: Quantifying geometric accuracy with unsupervised machine learning: using self-organizing map on fused filament fabrication additive manufacturing parts publication-title: J. Manuf. Sci. Eng. Trans. ASME – volume: 5 start-page: 277 year: 2020 end-page: 285 ident: bb0165 article-title: A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring publication-title: Prog Addit Manuf – volume: 128 year: 2019 ident: bb0065 article-title: High cycle fatigue life prediction of laser additive manufactured stainless steel: a machine learning approach publication-title: Int J Fatigue – start-page: 117 year: 2015 ident: bb0025 article-title: Study of physical phenomena of selective laser melting towards increased productivity publication-title: Thesis EPFL – volume: 554 start-page: 1828 year: 2013 end-page: 1834 ident: bb0035 article-title: Investigation of residual stresses induced during the selective laser melting process publication-title: Key Eng Mater – volume: 32 year: 2020 ident: bb0250 article-title: On the effect of shielding gas flow on porosity and melt pool geometry in laser powder bed fusion additive manufacturing publication-title: Addit Manuf – volume: 834 start-page: 93 year: 2015 end-page: 102 ident: bb0085 article-title: Determination of true temperature in selective laser melting of metal powder using infrared camera publication-title: Mater Sci Forum – volume: 9 start-page: 55214 year: 2021 end-page: 55231 ident: bb0115 article-title: Improving build quality in laser powder bed fusion using high dynamic range imaging and model-based reinforcement learning publication-title: IEEE Access – volume: 21 start-page: 517 year: 2018 end-page: 528 ident: bb0055 article-title: Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging publication-title: Addit Manuf – volume: 231 start-page: 488 year: 2016 end-page: 500 ident: bb0215 article-title: In situ morphology-based defect detection of selective laser melting through inline coherent imaging publication-title: J Mater Process Technol – volume: 21 start-page: 1 year: 2020 end-page: 13 ident: bb0245 article-title: The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation publication-title: BMC Genomics – volume: 212 start-page: 355 year: 2012 end-page: 364 ident: bb0255 article-title: Gas flow effects on selective laser melting (SLM) manufacturing performance publication-title: J Mater Process Technol – volume: 25 start-page: 151 year: 2019 end-page: 165 ident: bb0060 article-title: Using machine learning to identify in-situ melt pool signatures indicative of flaw formation in a laser powder bed fusion additive manufacturing process publication-title: Addit Manuf – year: 2021 ident: bb0005 article-title: Metal additive manufacturing – volume: 68 start-page: 451 year: 2019 end-page: 454 ident: bb0150 article-title: Machine learning-based image processing for on-line defect recognition in additive manufacturing publication-title: CIRP Ann – volume: 136 start-page: 445 year: 2019 end-page: 453 ident: bb0185 article-title: Experimental study of the process failure diagnosis in additive manufacturing based on acoustic emission publication-title: Meas J Int Meas Confed – volume: 95 start-page: 431 year: Apr. 2016 end-page: 445 ident: bb0015 article-title: Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing publication-title: Mater. Des. – volume: 51 start-page: 437 year: 2019 end-page: 455 ident: bb0170 article-title: In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes publication-title: IISE Trans. – year: 1999 ident: bb0230 article-title: Chapter 9: self-organizing maps publication-title: Neural Networks A Compr. Found – start-page: 1363 year: 2017 end-page: 1381 ident: bb0125 article-title: Machine learning for defect detection for PBFAm using high resolution layerwise imaging coupled with post-build CT scans publication-title: Solid freeform fabrication 2017: proceedings of the 28th annual international solid freeform fabrication symposium - an additive manufacturing conference, SFF 2017 – year: 2018 ident: bb0195 article-title: In situ and real-time monitoring of powder-bed AM by Combining acoustic emission and artificial intelligence publication-title: Industrializing additive manufacturing - proceedings of additive manufacturing in products and applications - AMPA2017 – volume: 5 start-page: 505 year: 2010 end-page: 514 ident: bb0050 article-title: Feedback control of layerwise laser melting using optical sensors publication-title: Phys. Procedia – volume: 15 start-page: 1 year: 2017 end-page: 11 ident: bb0040 article-title: Flaw detection in powder bed fusion using optical imaging publication-title: Addit. Manuf. – volume: 2018 year: 2018 ident: bb0190 article-title: Reinforcement learning: An Introduction – volume: 27 start-page: 42 year: 2019 end-page: 53 ident: bb0100 article-title: Automatic fault detection for laser powder-bed fusion using semi-supervised machine learning publication-title: Addit Manuf – year: 2013 ident: bb0205 article-title: Generation and detection of defects in metallic parts fabricated by selective laser melting and electron beam melting and their effects on mechanical properties publication-title: Solid Free. Fabr. – year: 2014 ident: bb0240 article-title: A novel self-organizing map (SOM) learning algorithm with nearest and farthest neurons publication-title: Alexandria Eng. J. – year: 2011 ident: bb0235 article-title: Information-theoretic approach to interpret internal representations of self-organizing maps publication-title: Self organizing maps - applications and novel algorithm design – volume: 156 start-page: 458 year: 2018 end-page: 469 ident: bb0105 article-title: Extraction and evaluation of melt pool, plume and spatter information for powder-bed fusion AM process monitoring publication-title: Mater Des – year: 2004 ident: bb0140 article-title: Laser cladding – volume: 20 year: 2019 ident: bb0155 article-title: Design rules and in-situ quality monitoring of thin-wall features made using laser powder bed fusion publication-title: Proceedings of the ASME 2019 14th international manufacturing science and engineering conference (MSEC2019) – volume: 50 year: 2018 ident: bb0095 article-title: Spatially weighted PCA for monitoring video image data with application to additive manufacturing publication-title: J. Qual. Technol. – start-page: 707 year: 2013 end-page: 712 ident: bb0145 article-title: High resolution imaging for inspection of Laser Beam Melting systems publication-title: 2013 IEEE international instrumentation and measurement technology conference (I2MTC) – volume: 49 start-page: 229 year: 2018 end-page: 239 ident: bb0045 article-title: In situ monitoring of selective laser melting of zinc powder via infrared imaging of the process plume publication-title: Robot. Comput. Integr. Manuf. – volume: 698505 year: 2008 ident: bb0020 article-title: Selective laser melting process monitoring with high-speed infra-red camera and pyrometer – volume: 209 start-page: 5973 year: 2009 end-page: 5980 ident: bb0160 article-title: Study on laser consolidation of metal powder with yb:fiber laser-evaluation of line consolidation structure publication-title: J Mater Process Technol – volume: 12 start-page: 683 year: 2011 end-page: 690 ident: bb0075 article-title: Design of an optical system for the in situ process monitoring of selective laser melting (SLM) publication-title: Phys. Procedia – volume: 19 start-page: 114 year: 2018 end-page: 126 ident: bb0070 article-title: Anomaly detection and classification in a laser powder bed additive manufacturing process using a trained computer vision algorithm publication-title: Addit Manuf – volume: 28 start-page: 666 year: Feb. 2019 end-page: 672 ident: bb0120 article-title: In situ quality monitoring in AM using acoustic emission: a reinforcement learning approach publication-title: J Mater Eng Perform – volume: 51 start-page: 29 year: 2019 end-page: 41 ident: bb0175 article-title: From in-situ monitoring toward high-throughput process control: cost-driven decision-making framework for laser-based additive manufacturing publication-title: J Manuf Syst – volume: 21 start-page: 598 year: 2018 end-page: 604 ident: bb0110 article-title: Acoustic emission for in situ quality monitoring in additive manufacturing using spectral convolutional neural networks publication-title: Addit Manuf – volume: 54 year: 2014 ident: bb0080 article-title: Process observation in fiber laser–based selective laser melting publication-title: Opt. Eng. – volume: 8 start-page: 316 year: 2008 end-page: 323 ident: bb0130 article-title: Prediction of laser solid freeform fabrication using neuro-fuzzy method publication-title: Appl Soft Comput J – volume: 114 start-page: 33 year: 2016 end-page: 42 ident: bb0200 article-title: Denudation of metal powder layers in laser powder bed fusion processes publication-title: Acta Mater – volume: 31 start-page: 375 year: 2020 end-page: 386 ident: bb0210 article-title: A deep neural network for classification of melt-pool images in metal additive manufacturing publication-title: J Intell Manuf – volume: 127 start-page: 4031 year: 2016 end-page: 4038 ident: bb0135 article-title: Analysis of laser powder deposition parameters: ANFIS modeling and ICA optimization publication-title: Optik (Stuttg) – volume: 52 start-page: 6110 year: 2014 end-page: 6135 ident: bb0090 article-title: Profile monitoring via sensor fusion: the use of PCA methods for multi-channel data publication-title: Int J Prod Res – volume: 96 start-page: 2791 year: 2018 end-page: 2801 ident: bb0030 article-title: Defect detection in selective laser melting technology by acoustic signals with deep belief networks publication-title: Int J Adv Manuf Technol – volume: 46 year: Oct. 2021 ident: bb0220 article-title: Development of a defect-detection platform using photodiode signals collected from the melt pool of laser powder-bed fusion publication-title: Addit Manuf – year: 2018 ident: bb0225 article-title: In-process monitoring systems for metal additive manufacturing publication-title: EOS Electro Optical Systems – volume: 136 year: 2014 ident: bb0010 article-title: A review on process monitoring and control in metal-based additive manufacturing publication-title: J Manuf Sci Eng – start-page: 707 year: 2013 ident: 10.1016/j.jmapro.2022.06.074_bb0145 article-title: High resolution imaging for inspection of Laser Beam Melting systems – volume: 15 start-page: 1 year: 2017 ident: 10.1016/j.jmapro.2022.06.074_bb0040 article-title: Flaw detection in powder bed fusion using optical imaging publication-title: Addit. Manuf. – volume: 96 start-page: 2791 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0030 article-title: Defect detection in selective laser melting technology by acoustic signals with deep belief networks publication-title: Int J Adv Manuf Technol doi: 10.1007/s00170-018-1728-0 – year: 2014 ident: 10.1016/j.jmapro.2022.06.074_bb0240 article-title: A novel self-organizing map (SOM) learning algorithm with nearest and farthest neurons publication-title: Alexandria Eng. J. doi: 10.1016/j.aej.2014.09.007 – volume: 32 year: 2020 ident: 10.1016/j.jmapro.2022.06.074_bb0250 article-title: On the effect of shielding gas flow on porosity and melt pool geometry in laser powder bed fusion additive manufacturing publication-title: Addit Manuf – volume: 51 start-page: 437 issue: 5 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0170 article-title: In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes publication-title: IISE Trans. doi: 10.1080/24725854.2017.1417656 – volume: 21 start-page: 598 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0110 article-title: Acoustic emission for in situ quality monitoring in additive manufacturing using spectral convolutional neural networks publication-title: Addit Manuf – start-page: 1363 year: 2017 ident: 10.1016/j.jmapro.2022.06.074_bb0125 article-title: Machine learning for defect detection for PBFAm using high resolution layerwise imaging coupled with post-build CT scans – volume: 156 start-page: 458 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0105 article-title: Extraction and evaluation of melt pool, plume and spatter information for powder-bed fusion AM process monitoring publication-title: Mater Des doi: 10.1016/j.matdes.2018.07.002 – volume: 127 start-page: 4031 issue: 8 year: 2016 ident: 10.1016/j.jmapro.2022.06.074_bb0135 article-title: Analysis of laser powder deposition parameters: ANFIS modeling and ICA optimization publication-title: Optik (Stuttg) doi: 10.1016/j.ijleo.2016.01.070 – volume: 231 start-page: 488 year: 2016 ident: 10.1016/j.jmapro.2022.06.074_bb0215 article-title: In situ morphology-based defect detection of selective laser melting through inline coherent imaging publication-title: J Mater Process Technol doi: 10.1016/j.jmatprotec.2015.12.024 – volume: 9 start-page: 55214 year: 2021 ident: 10.1016/j.jmapro.2022.06.074_bb0115 article-title: Improving build quality in laser powder bed fusion using high dynamic range imaging and model-based reinforcement learning publication-title: IEEE Access doi: 10.1109/ACCESS.2021.3067302 – year: 2013 ident: 10.1016/j.jmapro.2022.06.074_bb0205 article-title: Generation and detection of defects in metallic parts fabricated by selective laser melting and electron beam melting and their effects on mechanical properties publication-title: Solid Free. Fabr. – volume: 136 start-page: 445 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0185 article-title: Experimental study of the process failure diagnosis in additive manufacturing based on acoustic emission publication-title: Meas J Int Meas Confed doi: 10.1016/j.measurement.2018.12.067 – volume: 54 issue: 1 year: 2014 ident: 10.1016/j.jmapro.2022.06.074_bb0080 article-title: Process observation in fiber laser–based selective laser melting publication-title: Opt. Eng. doi: 10.1117/1.OE.54.1.011008 – year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0195 article-title: In situ and real-time monitoring of powder-bed AM by Combining acoustic emission and artificial intelligence – volume: 5 start-page: 277 year: 2020 ident: 10.1016/j.jmapro.2022.06.074_bb0165 article-title: A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring publication-title: Prog Addit Manuf doi: 10.1007/s40964-019-00108-3 – volume: 698505 issue: January 2008 year: 2008 ident: 10.1016/j.jmapro.2022.06.074_bb0020 article-title: Selective laser melting process monitoring with high-speed infra-red camera and pyrometer – volume: 68 start-page: 451 issue: 1 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0150 article-title: Machine learning-based image processing for on-line defect recognition in additive manufacturing publication-title: CIRP Ann doi: 10.1016/j.cirp.2019.03.021 – volume: 212 start-page: 355 issue: 2 year: 2012 ident: 10.1016/j.jmapro.2022.06.074_bb0255 article-title: Gas flow effects on selective laser melting (SLM) manufacturing performance publication-title: J Mater Process Technol doi: 10.1016/j.jmatprotec.2011.09.020 – volume: 554 start-page: 1828 year: 2013 ident: 10.1016/j.jmapro.2022.06.074_bb0035 article-title: Investigation of residual stresses induced during the selective laser melting process publication-title: Key Eng Mater doi: 10.4028/www.scientific.net/KEM.554-557.1828 – volume: 5 start-page: 505 issue: PART 2 year: 2010 ident: 10.1016/j.jmapro.2022.06.074_bb0050 article-title: Feedback control of layerwise laser melting using optical sensors publication-title: Phys. Procedia doi: 10.1016/j.phpro.2010.08.078 – volume: 31 start-page: 375 year: 2020 ident: 10.1016/j.jmapro.2022.06.074_bb0210 article-title: A deep neural network for classification of melt-pool images in metal additive manufacturing publication-title: J Intell Manuf doi: 10.1007/s10845-018-1451-6 – volume: 21 start-page: 517 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0055 article-title: Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging publication-title: Addit Manuf – year: 2021 ident: 10.1016/j.jmapro.2022.06.074_bb0005 – volume: 19 start-page: 114 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0070 article-title: Anomaly detection and classification in a laser powder bed additive manufacturing process using a trained computer vision algorithm publication-title: Addit Manuf – volume: 128 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0065 article-title: High cycle fatigue life prediction of laser additive manufactured stainless steel: a machine learning approach publication-title: Int J Fatigue doi: 10.1016/j.ijfatigue.2019.105194 – volume: 51 start-page: 29 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0175 article-title: From in-situ monitoring toward high-throughput process control: cost-driven decision-making framework for laser-based additive manufacturing publication-title: J Manuf Syst doi: 10.1016/j.jmsy.2019.02.005 – volume: 114 start-page: 33 year: 2016 ident: 10.1016/j.jmapro.2022.06.074_bb0200 article-title: Denudation of metal powder layers in laser powder bed fusion processes publication-title: Acta Mater doi: 10.1016/j.actamat.2016.05.017 – volume: 46 year: 2021 ident: 10.1016/j.jmapro.2022.06.074_bb0220 article-title: Development of a defect-detection platform using photodiode signals collected from the melt pool of laser powder-bed fusion publication-title: Addit Manuf – year: 2011 ident: 10.1016/j.jmapro.2022.06.074_bb0235 article-title: Information-theoretic approach to interpret internal representations of self-organizing maps – volume: 834 start-page: 93 issue: November year: 2015 ident: 10.1016/j.jmapro.2022.06.074_bb0085 article-title: Determination of true temperature in selective laser melting of metal powder using infrared camera publication-title: Mater Sci Forum doi: 10.4028/www.scientific.net/MSF.834.93 – volume: 52 start-page: 6110 issue: 20 year: 2014 ident: 10.1016/j.jmapro.2022.06.074_bb0090 article-title: Profile monitoring via sensor fusion: the use of PCA methods for multi-channel data publication-title: Int J Prod Res doi: 10.1080/00207543.2014.916431 – volume: 8 start-page: 316 issue: 1 year: 2008 ident: 10.1016/j.jmapro.2022.06.074_bb0130 article-title: Prediction of laser solid freeform fabrication using neuro-fuzzy method publication-title: Appl Soft Comput J doi: 10.1016/j.asoc.2007.02.002 – year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0225 article-title: In-process monitoring systems for metal additive manufacturing publication-title: EOS Electro Optical Systems – volume: 21 start-page: 1 issue: 1 year: 2020 ident: 10.1016/j.jmapro.2022.06.074_bb0245 article-title: The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation publication-title: BMC Genomics doi: 10.1186/s12864-019-6413-7 – volume: 49 start-page: 229 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0045 article-title: In situ monitoring of selective laser melting of zinc powder via infrared imaging of the process plume publication-title: Robot. Comput. Integr. Manuf. doi: 10.1016/j.rcim.2017.07.001 – year: 1999 ident: 10.1016/j.jmapro.2022.06.074_bb0230 article-title: Chapter 9: self-organizing maps – volume: 95 start-page: 431 year: 2016 ident: 10.1016/j.jmapro.2022.06.074_bb0015 article-title: Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing publication-title: Mater. Des. doi: 10.1016/j.matdes.2016.01.099 – volume: 20 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0155 article-title: Design rules and in-situ quality monitoring of thin-wall features made using laser powder bed fusion – start-page: 117 year: 2015 ident: 10.1016/j.jmapro.2022.06.074_bb0025 article-title: Study of physical phenomena of selective laser melting towards increased productivity – volume: 209 start-page: 5973 issue: 18–19 year: 2009 ident: 10.1016/j.jmapro.2022.06.074_bb0160 article-title: Study on laser consolidation of metal powder with yb:fiber laser-evaluation of line consolidation structure publication-title: J Mater Process Technol doi: 10.1016/j.jmatprotec.2009.07.017 – year: 2004 ident: 10.1016/j.jmapro.2022.06.074_bb0140 – volume: 12 start-page: 683 year: 2011 ident: 10.1016/j.jmapro.2022.06.074_bb0075 article-title: Design of an optical system for the in situ process monitoring of selective laser melting (SLM) publication-title: Phys. Procedia doi: 10.1016/j.phpro.2011.03.085 – volume: 28 start-page: 666 issue: 2 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0120 article-title: In situ quality monitoring in AM using acoustic emission: a reinforcement learning approach publication-title: J Mater Eng Perform doi: 10.1007/s11665-018-3690-2 – volume: 50 issue: 4 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0095 article-title: Spatially weighted PCA for monitoring video image data with application to additive manufacturing publication-title: J. Qual. Technol. doi: 10.1080/00224065.2018.1507563 – volume: 2018 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0190 – volume: 25 start-page: 151 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0060 article-title: Using machine learning to identify in-situ melt pool signatures indicative of flaw formation in a laser powder bed fusion additive manufacturing process publication-title: Addit Manuf – volume: 136 issue: 6 year: 2014 ident: 10.1016/j.jmapro.2022.06.074_bb0010 article-title: A review on process monitoring and control in metal-based additive manufacturing publication-title: J Manuf Sci Eng doi: 10.1115/1.4028540 – volume: 27 start-page: 42 year: 2019 ident: 10.1016/j.jmapro.2022.06.074_bb0100 article-title: Automatic fault detection for laser powder-bed fusion using semi-supervised machine learning publication-title: Addit Manuf – volume: 140 issue: 3 year: 2018 ident: 10.1016/j.jmapro.2022.06.074_bb0180 article-title: Quantifying geometric accuracy with unsupervised machine learning: using self-organizing map on fused filament fabrication additive manufacturing parts publication-title: J. Manuf. Sci. Eng. Trans. ASME doi: 10.1115/1.4038598 |
| SSID | ssj0012401 |
| Score | 2.4665303 |
| Snippet | This research studies the development of a machine-learning algorithm to detect porosity induced during laser powder bed fusion (LPBF) due to the lack of... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 476 |
| SubjectTerms | Additive manufacturing Clustering algorithm In-situ sensors Laser powder bed fusion Machine learning Quality assurance |
| Title | An unsupervised machine learning algorithm for in-situ defect-detection in laser powder-bed fusion |
| URI | https://dx.doi.org/10.1016/j.jmapro.2022.06.074 |
| Volume | 81 |
| WOSCitedRecordID | wos000878519800002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 2212-4616 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0012401 issn: 1526-6125 databaseCode: AIEXJ dateStart: 19990101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lj9MwELZKlwMcEE-xvOQDt8qrJs7WybFCRYDQikNBvUWO42zTbd0oj2X3Z_CPGcd2Em3Rwh64RFVaO0nny_iz_c0MQu8935fpzBNkKsKMBL44JREwUzKTOsePrvTQ7p7_-MrOzsLVKvo2Gv1ysTCXW6ZUeHUVFf_V1HAOjK1DZ-9g7q5TOAGfwehwBLPD8Z8MP1eTRlVNoZ1ABXRy18olpasPcT7h2_N9mdfrXSsxzBWp8rqZpFILO0gqaymcABKYtSwnxf5nKkuSQF9ZUzk7HhLaHVeNjpMwgY-FiUDoJYpLDvC4WJsaUlrFwa_3vQZgkVcOPoN8B_0a-DXcyYVtu1g7JZFdroCZrtNjdR7WByR4JtrZuWBTtcX60IDNBsNxYCoMHXh6s-iwOdnsODzQib6UScQa9COb282_MeB1MkSncNvEppdY9xJroR8L7qEjn51G4RgdzT8vVl-6rSmgQCYJr30OF4_ZigYP7-bPfGfAYZaP0SNrKzw3oHmCRlI9RQ8HKSmfoWSu8BA-2MIHO_jgDj4Y4IMtfPBN-MAXuIUP7uGDDXyeo-8fF8sPn4itw0EEvLY1CX3Jaap3cOk0lVIkVGRCwERH88fU87gPnJMLj3IWJsLLBKeZlCFPkiCbUsroCzRWeyVfIpzSUIS-iBifJgHjERcwuYuCWSZTP0q4OEbU_VuxsEnqda2UbXybrY4R6VoVJknLX37PnCFiSzQNgYwBXbe2fHXHK71GD_qX4A0a12Uj36L74rLOq_KdhdZvbM6oBg |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=An+unsupervised+machine+learning+algorithm+for+in-situ+defect-detection+in+laser+powder-bed+fusion&rft.jtitle=Journal+of+manufacturing+processes&rft.au=Taherkhani%2C+Katayoon&rft.au=Eischer%2C+Christopher&rft.au=Toyserkani%2C+Ehsan&rft.date=2022-09-01&rft.issn=1526-6125&rft.volume=81&rft.spage=476&rft.epage=489&rft_id=info:doi/10.1016%2Fj.jmapro.2022.06.074&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_jmapro_2022_06_074 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1526-6125&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1526-6125&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1526-6125&client=summon |