Options
CAD-free adjoint shape optimization of floating vessels
Citation Link: https://doi.org/10.15480/882.9233
Publikationstyp
Doctoral Thesis
Date Issued
2024
Sprache
English
Author(s)
Advisor
Referee
Title Granting Institution
Technische Universität Hamburg
Place of Title Granting Institution
Hamburg
Examination Date
2023-12-08
Institute
TORE-DOI
Citation
Technische Universität Hamburg (2024)
This thesis aims to improve fluid-dynamic shape optimization methods for non-parameterized (CAD-free) shapes. The focus is computing shape transformations for shapes with Lipschitz boundary and approximating the direction of the steepest descent in Banach spaces with the help of shape sensitivity information, which is calculated using an adjoint-based method. Further, considering geometric constraints and their algorithmic treatment is a central aspect of the thesis. The work investigates a p-Laplace relaxation to approximate Lipschitz transformations with the help of W-1-p functions. Further attempts are made to improve the approximation of the steepest descent direction in W-1-∞ using an approach based on the ADMM algorithm. The suggested approaches are applied in multiple numerical experiments, which successively deform a domain by point-wise movement of the computational grids.
Subjects
Shape optimization
Adjoint fluid dynamics
CFD
Steepest descent
Banach space
p-Laplace
DDC Class
620: Engineering
Funding Organisations
Loading...
Name
Peter_Marvin_Mueller_CAD-free_Adjoint_Shape_Optimization_of_Floating_Vessels.pdf
Size
25.06 MB
Format
Adobe PDF