Data organization in laser-based powder bed fusion for metals

Published

September 16, 2022

Author(s)

Shaw C. Feng, Albert T. Jones, Shengyen Li, Mostafa Yakout

Abstract

Data analytics (DA) and artificial intelligence (AI) have been chosen as the technologies for extracting new knowledge and making better decisions in additive manufacturing (AM) processes. They have been chosen because accurate and complete physics-based, process-simulation or mathematical models do not exist. DA and AI models should be based on measurable data collected by part and process sensors. This paper is focused on how to organize the data collected from such sensors. An example is also provided to show how to store AM-related data in a hierarchical data structure that is consistent with the data from multiple sensors. The data supports functions and properties at various stages in a product lifecycle. The associated metadata for both functions and properties are organized in the same hierarchical structure according to the relationships of machine, build, melting laser beams, process planning, in-situ monitoring, ex-situ inspection, material microstructure imaging, and mechanical testing. Sample data with meta data are stored in a file in the format of Hierarchical Data Format 5 (HDF5). The paper provides an organization of complex AM data that can support AM software tools for a variety of product lifecycle activities.

Proceedings Title

Manufacturing Letters

Conference Dates

June 27-July 1, 2022

Conference Location

West Lafayette, IN, US

Conference Title

50th SME North American Manufacturing Research Conference

Pub Type

Conferences

Download Paper

https://doi.org/10.1016/j.mfglet.2022.07.075

Local Download

Keywords

additive manufacturing, data organization, hierarchical data structure, sensor datasets

Citation

Feng, S. , Jones, A. , Li, S. and Yakout, M. (2022), Data organization in laser-based powder bed fusion for metals, Manufacturing Letters, West Lafayette, IN, US, [online], https://doi.org/10.1016/j.mfglet.2022.07.075, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=933853 (Accessed October 10, 2025)