國立虎尾科技大學 |

Applied Machine Learning Techniques for Additive Manufacturing Processes : = Toward in-situ Monitoring.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Applied Machine Learning Techniques for Additive Manufacturing Processes :/
Reminder of title:	Toward in-situ Monitoring.
Author:	Lucien, Charissa E.
Description:	1 online resource (101 pages)
Notes:	Source: Masters Abstracts International, Volume: 82-07.
Contained By:	Masters Abstracts International82-07.
Subject:	Industrial engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9798557036207

Applied Machine Learning Techniques for Additive Manufacturing Processes : = Toward in-situ Monitoring.
Lucien, Charissa E.

Applied Machine Learning Techniques for Additive Manufacturing Processes :Toward in-situ Monitoring. - 1 online resource (101 pages)

Source: Masters Abstracts International, Volume: 82-07.

Thesis (M.S.I.E.)--Florida Agricultural and Mechanical University, 2020.

Includes bibliographical references

Extrusion-based Additive Manufacturing (AM) is gaining attention from academia and industry due to its unique advantages compared to traditional manufacturing processes. Part quality inspection and the in-situ monitoring of those created parts is needed to play a critical role in process validation-an increasing requirement for full industrial adoption. The initial deposition has the most significant effect towards the quality of an AM produced composite part; however, there are +100 failure modes that stem from first layer issues. The lack of in-situ process analysis applied to digital manufacturing highlights the deficiencies of traditional monitoring and inspection. Traditional inspection processes have relied on the post-manual recognition of defects, which allows for errors of low efficiency or potential bias. This study presents 1) novel methods of in-situ process imaging and 2) application of a neural network (NN) approach towards AM quality inspection of the 1st-layer, by inspecting lack of fusion in 3D filament deposition. To obtain an appropriate model, experiments were performed on a variety of substrates and evaluated using a supervised machine learning NN algorithm which analyzed thermal imaging and in-situ printing images. The impact of the NN was then evaluated by observing the accuracy of the model framework during the training, testing, and validation of the datasets. The final developed NN models achieved final classification accuracies of ~96% and were then able to assist in defect-recognition in the AM industry. This Master's Thesis emphasizes the interdisciplinary nature of 3D printing and data science (i.e. materials-to-manufacturing) as a toolset of the Industrial & Manufacturing Engineer.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798557036207Subjects--Topical Terms:

679492
Industrial engineering.
Subjects--Index Terms:

Additive manufacturingIndex Terms--Genre/Form:

554714
Electronic books.

Applied Machine Learning Techniques for Additive Manufacturing Processes : = Toward in-situ Monitoring.
LDR:03178ntm a22004097 4500 001 1144209
005 20240531083759.5
006 m o d
007 cr mn ---uuuuu
008 250605s2020 xx obm 000 0 eng d
020 $a 9798557036207
035 $a (MiAaPQ)AAI28260027
035 $a AAI28260027
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Lucien, Charissa E. $3 1469159
245 1 0 $a Applied Machine Learning Techniques for Additive Manufacturing Processes : $b Toward in-situ Monitoring.
264 0 $c 2020
300 $a 1 online resource (101 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 82-07.
500 $a Advisor: Dickens, Tarik.
502 $a Thesis (M.S.I.E.)--Florida Agricultural and Mechanical University, 2020.
504 $a Includes bibliographical references
520 $a Extrusion-based Additive Manufacturing (AM) is gaining attention from academia and industry due to its unique advantages compared to traditional manufacturing processes. Part quality inspection and the in-situ monitoring of those created parts is needed to play a critical role in process validation-an increasing requirement for full industrial adoption. The initial deposition has the most significant effect towards the quality of an AM produced composite part; however, there are +100 failure modes that stem from first layer issues. The lack of in-situ process analysis applied to digital manufacturing highlights the deficiencies of traditional monitoring and inspection. Traditional inspection processes have relied on the post-manual recognition of defects, which allows for errors of low efficiency or potential bias. This study presents 1) novel methods of in-situ process imaging and 2) application of a neural network (NN) approach towards AM quality inspection of the 1st-layer, by inspecting lack of fusion in 3D filament deposition. To obtain an appropriate model, experiments were performed on a variety of substrates and evaluated using a supervised machine learning NN algorithm which analyzed thermal imaging and in-situ printing images. The impact of the NN was then evaluated by observing the accuracy of the model framework during the training, testing, and validation of the datasets. The final developed NN models achieved final classification accuracies of ~96% and were then able to assist in defect-recognition in the AM industry. This Master's Thesis emphasizes the interdisciplinary nature of 3D printing and data science (i.e. materials-to-manufacturing) as a toolset of the Industrial & Manufacturing Engineer.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Industrial engineering. $3 679492
650 4 $a Artificial intelligence. $3 559380
650 4 $a Computer science. $3 573171
653 $a Additive manufacturing
653 $a Adhesion
653 $a Machine learning
653 $a Material extrusion
653 $a Neural networks
653 $a Smart manufacturing
655 7 $a Electronic books. $2 local $3 554714
690 $a 0546
690 $a 0800
690 $a 0984
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Florida Agricultural and Mechanical University. $b Industrial and Manufacturing Engineering. $3 1469160
773 0 $t Masters Abstracts International $g 82-07.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28260027 $z click for full text (PQDT)