skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Machine-Learning-Based Monitoring of Laser Powder Bed Fusion

Abstract

A two-step machine learning approach to monitoring Laser Powder Bed Fusion (LPBF) additive manufacturing is demonstrated that enables on-the-fly assessments of laser track welds. First, in situ video melt pool data acquired during LPBF is labeled according to the (1) average and (2) standard deviation of individual track width and also (3) whether or not the track is continuous, measured post-build through an ex situ height map analysis algorithm. This procedure generates three ground truth labeled datasets for supervised machine learning. Using a portion of the labeled 10-millisecond video clips, a single Convolutional Neural Network architecture is trained to generate three distinct networks. With the remaining in situ LPBF data, the trained neural networks are tested and evaluated and found to predict track width, standard deviation, and continuity without the need for ex situ measurements. This two-step approach should benefit any LPBF system – or any additive manufacturing technology – where height-map-derived properties can serve as useful labels for in situ sensor data.

Authors:
 [1];  [1];  [1];  [1];  [1];  [1];  [1]; ORCiD logo [1]
  1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1481059
Alternate Identifier(s):
OSTI ID: 1468826
Report Number(s):
LLNL-JRNL-748383
Journal ID: ISSN 2365-709X; 933429
Grant/Contract Number:  
AC52-07NA27344
Resource Type:
Accepted Manuscript
Journal Name:
Advanced Materials Technologies
Additional Journal Information:
Journal Volume: 3; Journal Issue: 12; Journal ID: ISSN 2365-709X
Publisher:
Wiley
Country of Publication:
United States
Language:
English
Subject:
42 ENGINEERING

Citation Formats

Yuan, Bodi, Guss, Gabriel M., Wilson, Aaron C., Hau-Riege, Stefan P., DePond, Phillip J., McMains, Sara, Matthews, Manyalibo J., and Giera, Brian. Machine-Learning-Based Monitoring of Laser Powder Bed Fusion. United States: N. p., 2018. Web. doi:10.1002/admt.201800136.
Yuan, Bodi, Guss, Gabriel M., Wilson, Aaron C., Hau-Riege, Stefan P., DePond, Phillip J., McMains, Sara, Matthews, Manyalibo J., & Giera, Brian. Machine-Learning-Based Monitoring of Laser Powder Bed Fusion. United States. doi:10.1002/admt.201800136.
Yuan, Bodi, Guss, Gabriel M., Wilson, Aaron C., Hau-Riege, Stefan P., DePond, Phillip J., McMains, Sara, Matthews, Manyalibo J., and Giera, Brian. Wed . "Machine-Learning-Based Monitoring of Laser Powder Bed Fusion". United States. doi:10.1002/admt.201800136. https://www.osti.gov/servlets/purl/1481059.
@article{osti_1481059,
title = {Machine-Learning-Based Monitoring of Laser Powder Bed Fusion},
author = {Yuan, Bodi and Guss, Gabriel M. and Wilson, Aaron C. and Hau-Riege, Stefan P. and DePond, Phillip J. and McMains, Sara and Matthews, Manyalibo J. and Giera, Brian},
abstractNote = {A two-step machine learning approach to monitoring Laser Powder Bed Fusion (LPBF) additive manufacturing is demonstrated that enables on-the-fly assessments of laser track welds. First, in situ video melt pool data acquired during LPBF is labeled according to the (1) average and (2) standard deviation of individual track width and also (3) whether or not the track is continuous, measured post-build through an ex situ height map analysis algorithm. This procedure generates three ground truth labeled datasets for supervised machine learning. Using a portion of the labeled 10-millisecond video clips, a single Convolutional Neural Network architecture is trained to generate three distinct networks. With the remaining in situ LPBF data, the trained neural networks are tested and evaluated and found to predict track width, standard deviation, and continuity without the need for ex situ measurements. This two-step approach should benefit any LPBF system – or any additive manufacturing technology – where height-map-derived properties can serve as useful labels for in situ sensor data.},
doi = {10.1002/admt.201800136},
journal = {Advanced Materials Technologies},
number = 12,
volume = 3,
place = {United States},
year = {2018},
month = {9}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 2 works
Citation information provided by
Web of Science

Figures / Tables:

Figure 1 Figure 1: Schematic of video capture system. Adjustable mirrors trace the path of a focal point shared between the co-aligned high-speed camera, illumination laser, and high-power laser. The camera records in situ video taken during each welding event.

Save / Share:

Works referenced in this record:

Single track formation in selective laser melting of metal powders
journal, September 2010


Multifractal Analysis of Image Profiles for the Characterization and Detection of Defects in Additive Manufacturing
journal, January 2018

  • Yao, Bing; Imani, Farhad; Sakpal, Aniket S.
  • Journal of Manufacturing Science and Engineering, Vol. 140, Issue 3
  • DOI: 10.1115/1.4037891

Deep learning
journal, May 2015

  • LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
  • Nature, Vol. 521, Issue 7553
  • DOI: 10.1038/nature14539

Parametric analysis of the selective laser melting process
journal, July 2007


Optical diagnostics of selective laser melting and monitoring of single-track formation
journal, January 2017


Elucidation of melt flows and spatter formation mechanisms during high power laser welding of pure titanium
journal, August 2015

  • Nakamura, Hiroshi; Kawahito, Yousuke; Nishimoto, Koji
  • Journal of Laser Applications, Vol. 27, Issue 3
  • DOI: 10.2351/1.4922383

Monitoring and Adaptive Control of Laser Processes
journal, January 2014


Additive manufacturing of metallic components – Process, structure and properties
journal, March 2018


Computer Vision and Machine Learning for Autonomous Characterization of AM Powder Feedstocks
journal, December 2016


Effect of Laser Power and Scan Speed on Melt Pool Characteristics of Commercially Pure Titanium (CP-Ti)
journal, June 2017

  • Kusuma, Chandrakanth; Ahmed, Sazzad H.; Mian, Ahsan
  • Journal of Materials Engineering and Performance, Vol. 26, Issue 7
  • DOI: 10.1007/s11665-017-2768-6

Metal powder absorptivity: modeling and experiment
journal, January 2016

  • Boley, C. D.; Mitchell, S. C.; Rubenchik, A. M.
  • Applied Optics, Vol. 55, Issue 23
  • DOI: 10.1364/AO.55.006496

Flaw detection in powder bed fusion using optical imaging
journal, May 2017


Experimental analysis of spatter generation and melt-pool behavior during the powder bed laser beam melting process
journal, January 2018


Gaussian process-based surrogate modeling framework for process planning in laser powder-bed fusion additive manufacturing of 316L stainless steel
journal, September 2017

  • Tapia, Gustavo; Khairallah, Saad; Matthews, Manyalibo
  • The International Journal of Advanced Manufacturing Technology, Vol. 94, Issue 9-12
  • DOI: 10.1007/s00170-017-1045-z

Mesoscopic simulation model of selective laser melting of stainless steel powder
journal, November 2014


Laser powder-bed fusion additive manufacturing: Physics of complex melt flow and formation mechanisms of pores, spatter, and denudation zones
journal, April 2016


ImageNet classification with deep convolutional neural networks
journal, May 2017

  • Krizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E.
  • Communications of the ACM, Vol. 60, Issue 6
  • DOI: 10.1145/3065386

Density of additively-manufactured, 316L SS parts using laser powder-bed fusion at powers up to 400 W
journal, May 2014

  • Kamath, Chandrika; El-dasher, Bassem; Gallegos, Gilbert F.
  • The International Journal of Advanced Manufacturing Technology, Vol. 74, Issue 1-4
  • DOI: 10.1007/s00170-014-5954-9

In situ monitoring of selective laser melting of zinc powder via infrared imaging of the process plume
journal, February 2018


Characterization of Metal Powders Used for Additive Manufacturing
journal, January 2014

  • Slotwinski, J. A.; Garboczi, E. J.; Stutzman, P. E.
  • Journal of Research of the National Institute of Standards and Technology, Vol. 119
  • DOI: 10.6028/jres.119.018

The Unreasonable Effectiveness of Data
journal, March 2009

  • Halevy, Alon; Norvig, Peter; Pereira, Fernando
  • IEEE Intelligent Systems, Vol. 24, Issue 2
  • DOI: 10.1109/MIS.2009.36

Metal vapor micro-jet controls material redistribution in laser powder bed fusion additive manufacturing
journal, June 2017


Defect generation and propagation mechanism during additive manufacturing by selective beam melting
journal, November 2014


Evaluation of Powder Layer Density for the Selective Laser Melting (SLM) Process
journal, January 2017


Optical In-Process Temperature Monitoring of Selective Laser Melting
journal, January 2013


Data mining and statistical inference in selective laser melting
journal, January 2016

  • Kamath, Chandrika
  • The International Journal of Advanced Manufacturing Technology, Vol. 86, Issue 5-8
  • DOI: 10.1007/s00170-015-8289-2

Quality control of laser- and powder bed-based Additive Manufacturing (AM) technologies
journal, January 2010


Review of selective laser melting: Materials and applications
journal, December 2015

  • Yap, C. Y.; Chua, C. K.; Dong, Z. L.
  • Applied Physics Reviews, Vol. 2, Issue 4
  • DOI: 10.1063/1.4935926

Integration of physically-based and data-driven approaches for thermal field prediction in additive manufacturing
journal, February 2018


A fuzzy pattern recognition based system for monitoring laser weld quality
journal, July 2001

  • Park, Hyunsung; Rhee, Sehun; Kim, Dongcheol
  • Measurement Science and Technology, Vol. 12, Issue 8
  • DOI: 10.1088/0957-0233/12/8/345

Determination of True Temperature in Selective Laser Melting of Metal Powder Using Infrared Camera
journal, November 2015


Metal Additive Manufacturing: A Review
journal, April 2014


A review on quality control in additive manufacturing
journal, April 2018

  • Kim, Hoejin; Lin, Yirong; Tseng, Tzu-Liang Bill
  • Rapid Prototyping Journal, Vol. 24, Issue 3
  • DOI: 10.1108/RPJ-03-2017-0048

Direct measurements of temperature-dependent laser absorptivity of metal powders
journal, January 2015


Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing
journal, April 2016


Quality assessment in laser welding: a critical review
journal, May 2017

  • Stavridis, John; Papacharalampopoulos, Alexios; Stavropoulos, Panagiotis
  • The International Journal of Advanced Manufacturing Technology, Vol. 94, Issue 5-8
  • DOI: 10.1007/s00170-017-0461-4

    Figures/Tables have been extracted from DOE-funded journal article accepted manuscripts.