Skip to main content
eScholarship
Open Access Publications from the University of California

UC San Diego

UC San Diego Electronic Theses and Dissertations bannerUC San Diego

Gradient-based design optimization using CAD-based parameterization

Abstract

Reconciling optimized designs with the original model can be a time-consuming process. This thesis presents a process for gradient-based optimization using CAD-based parameterization that avoids the reconciliation process by directly altering CAD parameters and automatically updating the design. However, the complexities surrounding most commercial CAD tools make analytically obtaining design sensitivities necessary for gradient-based optimization virtually impossible. The value of this process lies in its ability to numerically compute design sensitivities.

This is done with the use of an intermediary surface discretization of the model called the master-mesh. The master-mesh maintains continuity as CAD parameters change through the use of a mesh smoothing process consisting of two optimizations. This smoothing process ensures that the master-mesh deforms smoothly with the design. Other meshes can be derived from the master-mesh, and therefore, can also deform smoothly with the design. This allows for the finite-difference method to be used to calculate their gradients.

This master-mesh is demonstrated with three distinct applications. The first application consists of an aerodynamic optimization of a wing designed with a custom geometry engine. The second application consists of an aerodynamic shape optimization of a jet plug nozzle designed in CATIA V5. The third application consists of an aerostructural optimization of a wing designed in FreeCAD.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View