Please use this identifier to cite or link to this item:
Title: Predicting build density in L-PBF through in-situ analysis of surface topography using powder bed scanner technology
Authors: Lu, Qingyang
Grasso, Marco
Le, Tan-Phuc
Seita, Matteo
Keywords: Engineering::Materials
Engineering::Mechanical engineering
Issue Date: 2022
Source: Lu, Q., Grasso, M., Le, T. & Seita, M. (2022). Predicting build density in L-PBF through in-situ analysis of surface topography using powder bed scanner technology. Additive Manufacturing, 51, 102626-.
Project: NRF-NRFF2018-05) 
Journal: Additive Manufacturing 
Abstract: The layerwise nature of additive manufacturing (AM) allows for in-situ monitoring of the consolidate material to identify defects on the fly and produce parts with improved reliability and performance. The main challenge in this paradigm, however, is that current methods have either limited measurement throughput or produce signals that are difficult to interpret and to relate to build properties. In this work, we present a new methodology that combines high-throughput in-situ measurements during laser powder bed fusion (L-PBF) with robust and unbiased numerical image analysis to predict build density from the surface topography of the consolidated material. The method relies on high resolution and large field of view optical scans of the layer—acquired through our powder bed scanner (PBS) technology—which we segment into “superpixels” to capture local and distributed differences in surface morphology and roughness. The high accuracy of our predictions together with the fast data acquisition and analysis enabled by the PBS and the low-dimensionality of the optical dataset after segmentation make our methodology an ideal candidate for in-line monitoring of materials produced by L-PBF. In addition, the ability to indirectly deduce a specific material property—namely density—as opposed to inferring a qualitative descriptor related to it makes our methodology unique and transferable to commercial powder bed fusion processes.
ISSN: 2214-7810
DOI: 10.1016/j.addma.2022.102626
Rights: © 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:MAE Journal Articles
MSE Journal Articles
SC3DP Journal Articles

Files in This Item:
File Description SizeFormat 
1-s2.0-S2214860422000343-main.pdf17.34 MBAdobe PDFThumbnail

Citations 50

Updated on Nov 26, 2022

Web of ScienceTM
Citations 50

Updated on Dec 1, 2022

Page view(s)

Updated on Dec 3, 2022


Updated on Dec 3, 2022

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.