Laser path optimization strategies for laser powder bed fusion
Abstract/Contents
- Abstract
- Metal additive manufacturing promises to enable the production of complex geometries with potentially exotic material properties. One leading paradigm of metal additive manufacturing is laser powder bed fusion, where a laser shines on a powder bed to selectively fuse powder together via heating, layer by layer, to build up a part. The path of the laser strongly affects the temperature field history of the part, which in turn determines the part's microstructure, as well as the presence of defects like porosity, warping, and residual stresses/strains. With the goal of elucidating how good or optimal laser paths can be identified, this thesis presents three main contributions. After an introduction and review of existing laser path optimization strategies in Chapter 1, Chapter 2 demonstrates the viability of modeling the optimal control problem as a board game and applying deep reinforcement learning a la AlphaGo Zero (albeit with major modifications) as an optimization technique. Then in Chapter 3, we show that by making the assumption that the effect of laser heating is finite in time, our optimal control problem can be modeled as an Traveling Salesperson Problem with History, which we prove can be transformed into equivalent Equality-Generalized Traveling Salesperson Problems and Traveling Salesperson Problems. Chapter 4 demonstrates how the structure of the Traveling Salesperson Problem with History can be used to develop a dynamic programming algorithm for solving and how it affects different solvers' performances. Finally, this thesis concludes with reflections on the potential and limitations of our work to improve laser powder bed fusion beyond computational models.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2023; ©2023 |
| Publication date | 2023; 2023 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Wang, Gradey |
|---|---|
| Degree supervisor | Darve, Eric |
| Degree supervisor | Lew, Adrian |
| Thesis advisor | Darve, Eric |
| Thesis advisor | Lew, Adrian |
| Thesis advisor | Kochenderfer, Mykel J, 1980- |
| Degree committee member | Kochenderfer, Mykel J, 1980- |
| Associated with | Stanford University, School of Engineering |
| Associated with | Stanford University, Department of Mechanical Engineering |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Gradey Wang. |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering. |
| Thesis | Thesis Ph.D. Stanford University 2023. |
| Location | https://purl.stanford.edu/rf959rw2791 |
Access conditions
- Copyright
- © 2023 by Gradey Wang
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...