Back to Search Start Over

An automated topology optimization interpreter that generates space frames.

Authors :
Myszka, David H.
Ives, Camden L.
Joo, James J.
Source :
Mechanics Based Design of Structures & Machines. 2024, Vol. 52 Issue 6, p3488-3510. 23p.
Publication Year :
2024

Abstract

This paper presents the methodology and theoretical foundation of topology optimization (TO) Results Spaceframe Interpreter, an automatic TO results interpreter that generates a closely associated space frame consisting of welded structural tubing or rectangular bars. TO is a computational technique that uses a finite element (FE) formulation to identify the most weight-efficient structure within a design domain. Density-based TO results in structures that take organic forms and is usually a tedious and cumbersome process to generate a computer-aided-design (CAD) model to manufacture through conventional techniques. The optimal topology frequently resembles a space frame, which is well-known as being a rigid, lightweight structure. The methodology of the TO results interpreter leverages several techniques from volumetric image processing and has four primary processes. First, the results are obtained from commercial FE/TO software and mapped into a cubic grid of voxels. Second, junction locations are extracted and member connectivity that represents a frame is identified. Third, a sizing optimization is incorporated to determine appropriate sectional dimensions of the circular or rectangular space frame members. Fourth, the optimized space frame geometry is imported into a CAD design tool to automatically create a design model. The logic and are implemented within the frame extraction processes. The automated TO interpreter is designed to interact with commercial FE analysis and CAD systems. The interpreter is demonstrated on various spatial examples including aerospace and automotive applications. In each case, the welded space frame closely resembles the TO result, with nearly equivalent stiffness and mass. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
15397734
Volume :
52
Issue :
6
Database :
Academic Search Index
Journal :
Mechanics Based Design of Structures & Machines
Publication Type :
Academic Journal
Accession number :
176934120
Full Text :
https://doi.org/10.1080/15397734.2023.2202234