Using autoencoded voxel patterns to predict part mass, required support material, and build time

Research output: Contribution to conferencePaper

Abstract

Additive Manufacturing (AM) allows designers to create intricate geometries that were once too complex or expensive to achieve through traditional manufacturing processes. Currently, designing parts using features specific to AM, commonly referred to as Design for Additive Manufacturing (DfAM), is restricted to experts in the field. As a result novices in industry may overlook potentially transformational design potential enabled by AM. This project aims to automate DfAM through deep learning making it accessible to a broader audience, and enabling designers of all skill levels to leverage unique AM geometries when creating new designs. To execute such an approach, a database of files was acquired from industry-sponsored AM challenges focused on lightweight design. These files were converted to a voxelized format, which provides more robust information for machine learning applications. Next, an autoencoder was constructed to a low-dimensional representation of the part designs. Finally, that autoencoder was used to construct a deep neural network capable of predicting various DfAM attributes. This work demonstrates a novel foray towards a more extensive DfAM support system that supports designers at all experience levels.

Original languageEnglish (US)
Pages1660-1674
Number of pages15
StatePublished - Jan 1 2020
Event29th Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference, SFF 2018 - Austin, United States
Duration: Aug 13 2018Aug 15 2018

Conference

Conference29th Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference, SFF 2018
CountryUnited States
CityAustin
Period8/13/188/15/18

All Science Journal Classification (ASJC) codes

  • Surfaces, Coatings and Films
  • Surfaces and Interfaces

Fingerprint Dive into the research topics of 'Using autoencoded voxel patterns to predict part mass, required support material, and build time'. Together they form a unique fingerprint.

  • Cite this

    Murphy, C., Meisel, N. A., Simpson, T. W., & McComb, C. (2020). Using autoencoded voxel patterns to predict part mass, required support material, and build time. 1660-1674. Paper presented at 29th Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference, SFF 2018, Austin, United States.