ORCID Number

0000-0001-9227-935X

Date of Award

Summer 7-8-2025

Embargo Period

10-12-2025

Access Type

Thesis - Open Access

Degree Name

Doctor of Philosophy in Aerospace Engineering

Department

Aerospace Engineering

Committee Chair

Ali Yeilaghi Tamijani ​

Committee Chair Email

tamijana@erau.edu

Committee Advisor

Ali Yeilaghi Tamijani ​

Committee Advisor Email

tamijana@erau.edu

Committee Co-Chair

Mandar Kulkarni

Committee Co-Chair Email

kulkarnm@erau.edu

First Committee Member

Sirish Namilae

First Committee Member Email

namilaes@erau.edu

Second Committee Member

Daewon Kim

Second Committee Member Email

kimd3c@erau.edu

Third Committee Member

Rafael Rodriguez

Third Committee Member Email

rodri7d6@erau.edu

College Dean

James W. Gregory

Abstract

This dissertation addresses two core challenges limiting the widespread application of Topology Optimization (TO): the difficulty in fabricating its complex designs, especially for Additive Manufacturing (AM), and its significant computational costs. It develops a unified design framework that directly embeds AM constraints such as overhang angles and build direction into robust TO formulations. To enhance manufacturability, two distinct methodologies are proposed. Firstly, a Solid Isotropic Material with Penalization (SIMP) framework introduces a three-stage robust optimization algorithm. Secondly, the Geometric Projection Topology Optimization (GPTO) method inherently integrates overhang constraints by controlling individual geometric components and their inclined angles relative to a rotating working plane. These frameworks consistently yield inherently self-supporting designs that maximize stiffness (minimize compliance) and maximize strength (assure stress limits), thereby minimizing material waste and post-processing while delivering high-performance components.

Concurrently, this work also directly confronts the high computational costs of TO, primarily stemming from intensive Finite Element Analysis (FEA) at each iteration. A physics-based Machine Learning (ML) framework is introduced to accelerate TO processes. This framework employs an offline, independent training strategy and utilizes a two-resolution setup. This approach reduces the computational cost by minimizing expensive fine-mesh FEA.

This research is crucial for bridging the gap between theoretical design optimality and practical AM feasibility, enabling scalable design of high-performance, manufacturable structure.

Download

Share

COinS