Parts internal structure definition using non-uniform patterned lattice optimization for mass reduction in additive manufacturing - Archive ouverte HAL
Article Dans Une Revue Engineering with Computers Année : 2018

Parts internal structure definition using non-uniform patterned lattice optimization for mass reduction in additive manufacturing

Résumé

Today, being able to generate and produce shapes that fit mechanical and functional requirements and having as low as possible mass is crucial for aerospace and automotive applications. Besides, the rise of new additive manufacturing technologies has widened the possibilities for designing and producing complex shapes and internal structures. However, current models, methods and tools still represent a limitation to that new horizon of printable shapes. This paper addresses the way internal lattice structures can be generated and optimized to reduce the mass of a product. A new framework is introduced that allows the modeling and optimization of non-uniform patterned lattice structures. Using non-uniform structures, additional degrees of freedom are introduced and allow the definition of a wide variety of shapes which can better fit the requirements. First, a non-uniform patterned lattice structure is generated using the results of an initial finite element analysis. This initial structure is then optimized while iteratively removing the beams considered as useless with respect to a user-specified mechanical criteria. At each iteration, the lattice structure is sent to a finite element solver that returns the von Mises stress map used to drive the simplification process. Here, the simulations are performed on the wireframe lattice structures to speed up the optimization loops. Once this process is completed, the final structure is no longer fully patterned, but it is reorganized to reduce the mass while satisfying the mechanical criteria. This approach is illustrated with examples coming from our prototype software.
Today, being able to generate and produce shapes that fit mechanical and functional requirements and having as low as possible mass is crucial for aerospace and automotive applications. Besides, the rise of new additive manufacturing technologies has widened the possibilities for designing and producing complex shapes and internal structures. However, current models, methods and tools still represent a limitation to that new horizon of printable shapes. This paper addresses the way internal lattice structures can be generated and optimized to reduce the mass of a product. A new framework is introduced that allows the modeling and optimization of non-uniform patterned lattice structures. Using non-uniform structures, additional degrees of freedom are introduced and allow the definition of a wide variety of shapes which can better fit the requirements. First, a non-uniform patterned lattice structure is generated using the results of an initial finite element analysis. This initial structure is then optimized while iteratively removing the beams considered as useless with respect to a user-specified mechanical criteria. At each iteration, the lattice structure is sent to a finite element solver that returns the von Mises stress map used to drive the simplification process. Here, the simulations are performed on the wireframe lattice structures to speed up the optimization loops. Once this process is completed, the final structure is no longer fully patterned, but it is re-organized to reduce the mass while satisfying the mechanical criteria. This approach is illustrated with examples coming from our prototype software.
Fichier principal
Vignette du fichier
LISPEN_EWC_2018_CHOUGRANI1.pdf (1.39 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01928580 , version 1 (21-12-2018)

Identifiants

  • HAL Id : hal-01928580 , version 1

Citer

Laurent Chougrani, Jean-Philippe Pernot, Philippe Véron, Stéphane Abed. Parts internal structure definition using non-uniform patterned lattice optimization for mass reduction in additive manufacturing. Engineering with Computers, 2018. ⟨hal-01928580⟩
31 Consultations
109 Téléchargements

Partager

More