Masters Theses

Keywords and Phrases

Deep Learning; Supports for Additive Manufacturing; Variational Autoencoder


"A key issue in metal additive manufacturing (AM) processes is the optimization of support geometry. Correct selection of support strategy can reduce build time, improve surface finish, reduce support removal time, and maximize build success. Strategies used to design support structure are time consuming and need skilled personnel. In this research we have deployed a deep generative model capable of making morphological modifications to the part. However, it is not similar to topology optimization where the aim is to reduce or eliminate support structure. The proposed model can make necessary changes to the part in order to transform it into a part with supports. Conventional shape synthesis algorithms synthesize new objects by retrieving and combining shapes and parts from a database. Presented in this research is a Variational Autoencoder (VAE) mapping the object space to a probabilistic latent space. Arithmetic is performed on the object vectors in the latent space to perform morphological operations on the geometry. This allows easy optimization of support geometry unlike conventional method of support generation. The dimensionally reduced latent space unlocks computationally tenable generation of variations in part geometry by learning features such as overhangs, holes, or internal voids which are absolutely necessary for support generation. This method will reduce human intervention and overall complexity in support design process and make the process quicker than before"-- Abstract, p. iii


Liou, Frank W.

Committee Member(s)

Sparks, Todd E.
Okafor, A. Chukwujekwu (Anthony Chukwujekwu)
Dagli, Cihan H., 1949-


Mechanical and Aerospace Engineering

Degree Name

M.S. in Manufacturing Engineering


Missouri University of Science and Technology

Publication Date

Spring 2021


viii, 32 pages

Note about bibliography

Includes bibliographical references (pages 29-31)


© 2021 Mugdha Swanand Joshi, All Rights Reserved

Document Type

Thesis - Open Access

File Type




Thesis Number

T 12146