Page 42 - MSAM-2-1

Materials Science in

Additive Manufacturing

ORIGINAL RESEARCH ARTICLE
Data imputation strategies for process

optimization of laser powder bed fusion of
Ti6Al4V using machine learning

2
1
Guo Dong Goh , Xi Huang , Sheng Huang , Jia Li Janessa Thong , Jia Jun Seah ,
3
3
3
Wai Yee Yeong 1,2,3 *
1 Singapore Centre for 3D Printing, School of Mechanical and Aerospace Engineering, Nanyang
Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore
2 HP-NTU Digital Manufacturing Corporate Lab, Nanyang Technological University, Singapore
3 School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang
Avenue, Singapore 639798, Singapore

Abstract

A database linking process parameters and material properties for additive
manufacturing enables the performance of the material to be determined based on
the process parameters, which are useful in the design and fabrication stage of a
product. The data, however, are often incomplete as each individual research work
focused on certain process parameters and material properties due to the wide range
of variables available. Imputation of missing data is thus required to complete the
material library. In this work, we attempt to collate the data of Ti6Al4V, a popular alloy
used in aerospace and biomedical industries, fabricated using powder bed fusion, or
*Corresponding author: commonly known as selective laser melting (SLM). Various imputation techniques
Wai Yee Yeong
(wyyeong@ntu.edu.sg) of missing data of the SLM Ti6Al4V dataset, such as the k-nearest neighbor (kNN),
multivariate imputation by chained equations, and graph imputation neural network
Citation: Goh GD, Huang X, (GINN) are investigated in this article. It was observed that kNN performed better in
Huang S, et al., 2023, Data
imputation strategies for process imputing variables related to process parameters, whereas GINN performed better in
optimization of laser powder bed variables related to material properties. To further improve the quality of imputation,
fusion of Ti6Al4V using machine a strategy to use the median of the imputed values obtained from the three models
learning. Mater Sci Add Manuf,
2(1): 50. has resulted in significant improvement in terms of the relative mean square error.
https://doi.org/10.36922/msam.50 Self-organizing map was used to visualize the relationship among the process
Received: February 1, 2023 parameters and the material properties.
Accepted: March 7, 2023
Keywords: Additive manufacturing; 3D printing; Selective laser melting; Powder bed
Published Online: March 22, 2023
fusion; Machine learning; Data analytics; Imputation
Copyright: © 2023 Author(s).
This is an Open Access article
distributed under the terms of the
Creative Commons Attribution
License, permitting distribution, 1. Introduction
and reproduction in any medium,
provided the original work is Ti6Al4V is one of the most popular titanium alloys given its excellent material properties,
properly cited. including high strength, low density, and high corrosion resistance, and is used in a wide
Publisher’s Note: AccScience variety of industries, such as in aerospace for aircraft components and in biomedical
Publishing remains neutral with for implants . Instead of using traditional manufacturing methods, selective laser
[1]
regard to jurisdictional claims in
published maps and institutional melting (SLM) of Ti6Al4V allows for more complex parts to be created. It is an additive
affiliations. manufacturing technique, categorized as powder bed fusion (PBF), which involves

Volume 2 Issue 1 (2023) 1 https://doi.org/10.36922/msam.50