Keywords

Additive manufacturing; Machine learning; Process-structure-property relationships

Abstract

Additive manufacturing (AM) has been increasingly used in the automotive, aerospace, defense, medical devices, and energy sectors, due to its ability to fabricate parts with complex geometries and excellent mechanical behavior. However, it is challenging to design for AM and optimize process parameters because the process-structure-property (PSP) relationships in AM are complex. This study investigates the PSP relationships in AM through experimentation and machine learning. First, the effect of process parameters on the flexural behavior of soft-hard sandwich beams fabricated by multi-material AM is investigated. How the flexural modulus, flexural strength, and failure mechanisms of the soft-hard sandwich beams relate to process parameters such as nozzle temperatures and print speeds for soft and hard materials is revealed. Second, a novel multi-process AM system is developed to fabricate thermoplastics and thermosets in a single print. The interfacial behavior between thermoplastics and thermosets, including their in-situ surface morphology, bonding strength, and failure mechanisms, is studied. Third, machine learning and feature representation approaches are developed to predict the stress-strain relationships of additively manufactured metamaterials with varying structure and process parameters. Two novel feature representations are developed to transform process and structure parameters from tabular format into images. Fourth, a transfer learning framework is introduced to predict the stress-strain relationships of additively manufactured metamaterials in the target domain with a small dataset by leveraging prior knowledge in the source domain.

Completion Date

2024

Semester

Summer

Committee Chair

Wu, Dazhong

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Department of Mechanical and Aerospace Engineering

Degree Program

Mechanical Engineering

Format

application/pdf

Identifier

DP0028874

URL

https://stars.library.ucf.edu/cgi/viewcontent.cgi?article=1390&context=etd2023

Language

English

Rights

In copyright

Release Date

2-15-2028

Length of Campus-only Access

3 years

Access Status

Doctoral Dissertation (Campus-only Access)

Campus Location

Orlando (Main) Campus

Accessibility Status

Meets minimum standards for ETDs/HUTs

Download

Restricted to the UCF community until 2-15-2028; it will then be open access.

Share

COinS