TY - JOUR
T1 - Mathematical morphology-based point cloud analysis techniques for geometry assessment of 3D printed concrete elements
AU - Nair, Sooraj A.O.
AU - Sant, Gaurav
AU - Neithalath, Narayanan
N1 - Funding Information:
The authors sincerely acknowledge the support from U.S. National Science Foundation ( CMMI: 1727445 ; OISE: 2020095 ) towards this project. The authors also acknowledge the support from Salt River Materials Group, Omya, and BASF in donating the materials, and the use of 3D printing and material characterization facilities within the Laboratory for the Science of Sustainable Infrastructural Materials (LS-SIM) at Arizona State University in the completion of this project.
Publisher Copyright:
© 2021
PY - 2022/1
Y1 - 2022/1
N2 - In 3D printing of cement-based materials, it is imperative to ensure geometrical consistency of the print with the as-designed/modeled system. Time-dependent, deformable systems like concrete present multiple challenges in ensuring appropriate post-print quality. This paper presents a suite of point cloud comparison techniques, which can be used individually or in combination, to quantify the amount of mismatch between the as-designed and as-printed systems, using morphological analysis. A semi-quantitative error distance method is proposed, which can be easily accomplished using direct mapping of the actual and reference point clouds. A print accuracy index (PAI) based on centroidal distances is proposed as a global quantifier of the print quality. Furthermore, a topological set theory (TST)-based approach is used to determine layer-wise overlap, which helps in isolating localized inconsistencies. The methods are tested on a variety of small cuboids, and further verified using a larger mortar print. It is expected that these methodologies can be suitably adapted to indicate the efficiency of the print, after the fact, or during printing. The latter facilitates in-line quality checks, that can in turn lead to real-time alterations in the materials or processes.
AB - In 3D printing of cement-based materials, it is imperative to ensure geometrical consistency of the print with the as-designed/modeled system. Time-dependent, deformable systems like concrete present multiple challenges in ensuring appropriate post-print quality. This paper presents a suite of point cloud comparison techniques, which can be used individually or in combination, to quantify the amount of mismatch between the as-designed and as-printed systems, using morphological analysis. A semi-quantitative error distance method is proposed, which can be easily accomplished using direct mapping of the actual and reference point clouds. A print accuracy index (PAI) based on centroidal distances is proposed as a global quantifier of the print quality. Furthermore, a topological set theory (TST)-based approach is used to determine layer-wise overlap, which helps in isolating localized inconsistencies. The methods are tested on a variety of small cuboids, and further verified using a larger mortar print. It is expected that these methodologies can be suitably adapted to indicate the efficiency of the print, after the fact, or during printing. The latter facilitates in-line quality checks, that can in turn lead to real-time alterations in the materials or processes.
KW - 3D printed concrete
KW - Error distance
KW - Mathematical morphology quality control
KW - Point cloud analysis
KW - Print accuracy index
UR - http://www.scopus.com/inward/record.url?scp=85120319329&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85120319329&partnerID=8YFLogxK
U2 - 10.1016/j.addma.2021.102499
DO - 10.1016/j.addma.2021.102499
M3 - Article
AN - SCOPUS:85120319329
VL - 49
JO - Additive Manufacturing
JF - Additive Manufacturing
SN - 2214-8604
M1 - 102499
ER -