elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Automatic Processing and Cross Section Analysis of Topology Optimization Results

Gomes Alves, Christian and Barthel, Yannick (2021) Automatic Processing and Cross Section Analysis of Topology Optimization Results. In: NAFEMS World Congress 2021. NAFEMS World Congress 2021, 25.-29.10.2021, online.

[img] PDF
2MB

Official URL: https://www.nafems.org/publications/resource_center/nwc21-191-c/

Abstract

Finite element-based topology optimizations have become a vital tool in the design of lightweight-optimized components and structures. Used in the early stages of the product development process, they can help identify load paths and shape the fundamental design of a product. However, the interpretation of topology optimization results can be a challenging and time-consuming task. In order to properly analyze the topologies, both knowledge in the field of finite element computations as well as computer-aided engineering and design are necessary. This multi-disciplinary engineering expertise can be expensive and not feasible for small and medium-sized businesses. By automating this vital step in a product development process, the utilization of topology optimizations in all product development processes can be greatly promoted. We propose a two-fold process for automatically processing density-based topology optimization results for future CAD design proposals: extracting a wireframe and deriving cross sections. The wireframe extraction supports mixed-element models featuring two-dimensional and three-dimensional finite elements and optional mirror symmetry consideration. The cross section extraction is currently available for three-dimensional finite elements. The wireframe extraction is fundamentally based on voxelization and skeletonization of a finite element- and density-based topology optimization. The cross section extraction makes use of the element density as well as the stress distributions from the topology optimization result. At first, load conditions are identified for each beam of the previously extracted wireframe and an appropriate beam profile is chosen from a set of profiles (currently circle and rectangle, both filled or hollow, and I-beams). Then, using a least square loss optimization and image processing-based shape averaging, the geometric dimensions of each selected beam profile are determined. In the end, a model is available which can be converted into a parametric CAD model for further design work.

Item URL in elib:https://elib.dlr.de/144577/
Document Type:Conference or Workshop Item (Speech)
Title:Automatic Processing and Cross Section Analysis of Topology Optimization Results
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Gomes Alves, ChristianChristian.GomesAlves (at) dlr.dehttps://orcid.org/0000-0002-0897-6951
Barthel, YannickUNSPECIFIEDUNSPECIFIED
Date:27 October 2021
Journal or Publication Title:NAFEMS World Congress 2021
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:No
In ISI Web of Science:No
Series Name:https://www.nafems.org/publications/resource_center/nwc21-191-b/
Status:Published
Keywords:Automation, Topology Optimization, Code Coupling, Interpreting Optimization Results, Democratisation
Event Title:NAFEMS World Congress 2021
Event Location:online
Event Type:international Conference
Event Dates:25.-29.10.2021
Organizer:NAFEMS
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Rail Transport
DLR - Research area:Transport
DLR - Program:V SC Schienenverkehr
DLR - Research theme (Project):V - NGT BIT
Location: Stuttgart
Institutes and Institutions:Institute of Vehicle Concepts > Vehicle Architectures and Lightweight Design Concepts
Deposited By: Gomes Alves, Christian
Deposited On:16 Dec 2021 15:24
Last Modified:26 Jan 2022 09:30

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.