Skip to Main content Skip to Navigation
Theses

Une méthode d'optimisation topologique multi-échelle CAO-compatible

Abstract : This Ph.D. thesis focuses on the integration of specificities of the multi-scale analysis and of the additive manufacturing (AM) process, into the topology optimisation (TO) algorithm based on the pseudo-density as topological descriptor and on the non-uniform rational basis spline (NURBS) hyper-surfaces developed at theI2M laboratory in Bordeaux. The goal is to facilitate the work of the designer during the different stages of the AM digital chain, by reducing the time required for each stage. Two main challenges have been faced in this Ph.D. thesis. Firstly, the development of a semi-automatic surface reconstruction strategy to recover the boundary of the optimised topology by minimising the computational resources (time, memory, etc.) dedicated to this task is proposed. Secondly, the integration of the peculiarities of multi-scale analyses in the TO process is faced. Regarding this aspect, a theoretical/numerical framework allowing the concurrent optimisation of thetopological descriptors defined at multiple scales of the problem at hand has been developed. In this background design requirements of different nature, as the manufactruing constraint on the minimum member size, the scale separation condition, lighteness, generalised compliance (in presence of mixed non-zero Neumann-Dirichlet boundary conditions) have been included in the problem formulation, by exploiting theproperties of the NURBS formalism. The effectiveness of the proposed multi-scale TO method has been tested on both 2D and 3D benchmark structures and validated via the results of experimental tests (three-point bending tests).
Complete list of metadata

https://pastel.archives-ouvertes.fr/tel-03741356
Contributor : ABES STAR :  Contact
Submitted on : Monday, August 1, 2022 - 10:56:12 AM
Last modification on : Friday, August 5, 2022 - 2:54:01 PM

File

106078_BERTOLINO_2022_archivag...
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03741356, version 1

Citation

Giulia Bertolino. Une méthode d'optimisation topologique multi-échelle CAO-compatible. Eco-conception. HESAM Université, 2022. Français. ⟨NNT : 2022HESAE035⟩. ⟨tel-03741356⟩

Share

Metrics

Record views

0

Files downloads

0