Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning

In powder bed fusion additive manufacturing, machines are often equipped with in-situ sensors to monitor the build environment as well as machine actuators and subsystems. The data from these sensors offer rich information about the consistency of the fabrication process within a build and across bu...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	William Halsey, Derek Rose, Luke Scime, Ryan Dehoff, Vincent Paquit
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	additive manufacturing (3D printing) machine learning spatio - temporal analysis explainable AI process-structure-property linkage defect detection Mechanical engineering and machinery TJ1-1570
Acceso en línea:	https://doaj.org/article/00ef93f7d9dd4865b96ad70a73562665
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:00ef93f7d9dd4865b96ad70a73562665
record_format	dspace
spelling	oai:doaj.org-article:00ef93f7d9dd4865b96ad70a735626652021-11-15T07:00:37ZLocalized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning2297-307910.3389/fmech.2021.767444https://doaj.org/article/00ef93f7d9dd4865b96ad70a735626652021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fmech.2021.767444/fullhttps://doaj.org/toc/2297-3079In powder bed fusion additive manufacturing, machines are often equipped with in-situ sensors to monitor the build environment as well as machine actuators and subsystems. The data from these sensors offer rich information about the consistency of the fabrication process within a build and across builds. This information may be used for process monitoring and defect detection; however, little has been done to leverage this data from the machines for more than just coarse-grained process monitoring. In this work we demonstrate how these inherently temporal data may be mapped spatially by leveraging scan path information. We then train a XGBoost machine learning model to predict localized defects—specifically soot–using only the mapped process data of builds from a laser powder bed fusion process as input features. The XGBoost model offers a feature importance metric that will help to elucidate possible relationships between the process data and observed defects. Finally, we analyze the model performance spatially and rationalize areas of greater and lesser performance.William HalseyWilliam HalseyDerek RoseLuke ScimeLuke ScimeRyan DehoffRyan DehoffVincent PaquitVincent PaquitFrontiers Media S.A.articleadditive manufacturing (3D printing)machine learningspatio - temporal analysisexplainable AIprocess-structure-property linkagedefect detectionMechanical engineering and machineryTJ1-1570ENFrontiers in Mechanical Engineering, Vol 7 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	additive manufacturing (3D printing) machine learning spatio - temporal analysis explainable AI process-structure-property linkage defect detection Mechanical engineering and machinery TJ1-1570
spellingShingle	additive manufacturing (3D printing) machine learning spatio - temporal analysis explainable AI process-structure-property linkage defect detection Mechanical engineering and machinery TJ1-1570 William Halsey William Halsey Derek Rose Luke Scime Luke Scime Ryan Dehoff Ryan Dehoff Vincent Paquit Vincent Paquit Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning
description	In powder bed fusion additive manufacturing, machines are often equipped with in-situ sensors to monitor the build environment as well as machine actuators and subsystems. The data from these sensors offer rich information about the consistency of the fabrication process within a build and across builds. This information may be used for process monitoring and defect detection; however, little has been done to leverage this data from the machines for more than just coarse-grained process monitoring. In this work we demonstrate how these inherently temporal data may be mapped spatially by leveraging scan path information. We then train a XGBoost machine learning model to predict localized defects—specifically soot–using only the mapped process data of builds from a laser powder bed fusion process as input features. The XGBoost model offers a feature importance metric that will help to elucidate possible relationships between the process data and observed defects. Finally, we analyze the model performance spatially and rationalize areas of greater and lesser performance.
format	article
author	William Halsey William Halsey Derek Rose Luke Scime Luke Scime Ryan Dehoff Ryan Dehoff Vincent Paquit Vincent Paquit
author_facet	William Halsey William Halsey Derek Rose Luke Scime Luke Scime Ryan Dehoff Ryan Dehoff Vincent Paquit Vincent Paquit
author_sort	William Halsey
title	Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning
title_short	Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning
title_full	Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning
title_fullStr	Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning
title_full_unstemmed	Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning
title_sort	localized defect detection from spatially mapped, in-situ process data with machine learning
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/00ef93f7d9dd4865b96ad70a73562665
work_keys_str_mv	AT williamhalsey localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT williamhalsey localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT derekrose localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT lukescime localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT lukescime localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT ryandehoff localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT ryandehoff localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT vincentpaquit localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning AT vincentpaquit localizeddefectdetectionfromspatiallymappedinsituprocessdatawithmachinelearning
_version_	1718428473814417408

Localized Defect Detection from Spatially Mapped, In-Situ Process Data With Machine Learning

Ejemplares similares