Predicting part distortion field in additive manufacturing: a data-driven framework

Osama Aljarrah (), Jun Li (), Alfa Heryudono (), Wenzhen Huang () and Jing Bi ()
Additional contact information
Osama Aljarrah: Youngstown State University
Jun Li: University of Massachusetts Dartmouth
Alfa Heryudono: University of Massachusetts Dartmouth
Wenzhen Huang: University of Massachusetts Dartmouth
Jing Bi: Dassault Systẽmes SIMULIA Corp

Journal of Intelligent Manufacturing, 2023, vol. 34, issue 4, No 25, 1975-1993

Abstract: Abstract The underlying complicated spatiotemporal thermo-mechanical processes in additive manufacturing (AM) technology pose challenges in predicting and optimizing the as-built part quality for production use. Physics-based simulations are being developed to provide reliable predictions such as part distortions, residual stresses/strains, microstructure contents, and grain morphology, which can guide the product design and manufacturing process for improved part quality. However, due to the complexity of the problem, e.g., multi-physics and multi-scale, physics-based models need the expertise to build an extensive simulation time preventing its use in real-time monitoring and control. This study proposes an inductive data-driven framework to simulate the part distortion field for additively manufactured components. The part distortion field data were collected from different AM build strategies using finite element (FE) simulations. The FE model consists of multi-layer part distortion fields as outputs constituted by AM process parameters as inputs. The surrogate model consists of two stages: (1) self-organizing map to project the high-dimensional spatial field of the part distortion into a likelihood estimator, (2) hybrid self-organizing methods to predict the extracted features and reconstruct the part distortion field. The data-driven model evaluates different build scenarios’ effects on the distortion field for additively manufactured parts. The results correlated well with FE simulations and established a prediction-based compensation strategy to reduce part distortions. It should be noted that while the framework is currently applied to FE simulation data, the integrated data-driven methods can also be deployed on experimentally measured 3D cloud data or other high throughput field measurements.

Keywords: Additive manufacturing; Part distortions; Data-driven model; Self-organizing map; Group of adaptive models evolution (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://link.springer.com/10.1007/s10845-021-01902-z Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:34:y:2023:i:4:d:10.1007_s10845-021-01902-z

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-021-01902-z

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().