An official website of the United States government
Here’s how you know
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
Machine Learning-driven Process-Structure-Property Analytical Framework for Additive Manufacturing
Published
Author(s)
Hyunwoong Ko, Zhuo Yang, Yande Ndiaye, Paul Witherell, Yan Lu
Abstract
Data analytics with Machine Learning (ML) and Artificial Intelligence (AI) offers high potential to continuously transform AM data to newfound knowledge of Process-Structure-Property (PSP) relationships. In AM, however, realizing the potential is still limited largely due to the lack of a systematic way to learn the PSP relationships for various AM processes. To address the limitation, this paper proposes a novel, ML-driven framework, which consists of three tiers: (1) knowledge of predictive models and physics, (2) features of interest, and (3) raw data. The framework defines a PSP-learning process with two sub-processes. The first uses a top-down, knowledge-graph-guided approach to generate the requirements for predictive analytics and data acquisition. The second uses a bottom-up, data-driven approach to model and construct new, PSP knowledge. Together, these processes connect the proposed framework to control-decisions and physical/virtual AM systems, respectively. The paper includes a case study based on Laser Powder Bed Fusion processes including AM Metrology Testbed at National Institute of Standards and Technology (NIST). The case study introduces predictive, PSP-ML models and PSP knowledge extracted from the models. We also demonstrate the framework using a ML-Integrated Knowledge Extraction module called MIKE in NIST's collaborative AM Material Database. The framework newly enables a systematic, hybrid, PSP-modeling approach for AM that can couple physics knowledge with the versatility of data-driven, ML models. Using the approach, the framework continuously updates the models (1) to improve understanding of dynamically generated AM data and (2) to link sub-models into coupled, PSP models. Based on the improved understanding, the framework facilitates control-decisions for AM at multiple scales.
Ko, H.
, Yang, Z.
, Ndiaye, Y.
, Witherell, P.
and Lu, Y.
(2023),
Machine Learning-driven Process-Structure-Property Analytical Framework for Additive Manufacturing, Journal of Manufacturing Systems, [online], https://doi.org/10.1016/j.jmsy.2022.09.010, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=932370
(Accessed October 9, 2024)