Virtual Optimization of 3D Printed Structures with Tailored Mechanical and Functional Properties

Research output: ThesisDoctoral Thesis

Abstract

Metamaterials are a class of artificially structured materials with special and often unique material properties due to their distinctive unit cell-based periodic architecture. The multiscale nature of metamaterials in combination with the strong link between material properties and unit cell geometry creates a high customization and optimization potential. The actual unit cell geometry, however, can be significantly influenced by the manufacturing process, due to their frequently very intricate, design. The possibility to predict the effect of the manufacturing process on the geometry is therefore vital for the design process and a comprehensive optimization of metamaterials. Their special characteristics lead to the necessity of new strategies tailored specifically towards the design process of metamaterials in order to exploit their full potential. This thesis presents two simulation-based frameworks to improve the efficiency of the entire design process, and to create the first steps towards a comprehensive optimization strategy for metamaterial parts. Due to the high potential for optimization and the need for an accurate prediction of geometric parameters after manufacturing, an optimization framework and a process simulation tool are developed. The optimization framework covers the optimization of the unit cell distribution in metamaterial components to achieve a predefined deformation behavior. The process simulation complements the optimization framework by the prediction of the manufacturing effects of the Digital Light Processing 3D printing method. The developed optimization framework combines numerical homogenization, stiffness tensor interpolation and Finite Element simulation-based black-box optimization. A separation into a material pre-processing part and the actual optimization part results in an improved optimization efficiency. Furthermore, the addition of an automated material section discretization routine reduces the dependency of the framework on the user-defined material section discretization, and improves the convergence behavior. Regarding the process simulation, a cure kinetics model, a degree of cure-dependent material model and an element activation routine are implemented into the Abaqus Finite Element software. Thereby, the framework facilitates the evaluation of effects such as residual stresses, warpage and print accuracy during the manufacturing process. The developed process simulation uses a modular setup to enable the ability to choose the level of detail, and the associated computational expense based on the user¿s needs. Additional modules take the effects of the uncured resin surrounding a part and the temperature evolution due to the exothermic reaction into account. The simulation-based frameworks developed as part of this thesis represent a successful implementation of custom tools to help with designing metamaterial components thus providing a starting point for further development of a comprehensive optimization strategy for supporting their design process.

Details

Translated title of the contributionVirtuelle Optimierung von 3D gedruckten Strukturen mit angepassten mechanischen und funktionalen Eigenschaften
Original languageEnglish
QualificationDr.mont.
Awarding Institution
Supervisors/Advisors
  • Elgeti, Stefanie, Assessor A (external), External person
  • Schuecker, Clara, Supervisor (internal)
  • Schlögl, Sandra, Co-Supervisor (external)
  • Muliana, Anastasia, Assessor B (external), External person
DOIs
Publication statusPublished - 2024