Long Jiang

Type

Text

Type

Dissertation

Advisor

Advisor: Chen, Shikui | Committee members: Ge, Jeffrey; Nakamura, Toshio; Jiao, Xiangmin

Date

2019

Keywords

Topology Optimization

Department

Department of Mechanical Engineering

Language

en

Source

This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.

Identifier

http://hdl.handle.net/11401/78894

Publisher

The Graduate School, Stony Brook University: Stony Brook, NY.

Format

application/pdf

Abstract

The design of structure is a research topic explored by human beings for hundreds of years. However, for a long period of time, the designing process is always an experience or inspiration based tinkering rather than a scientifically based technology. As the designing requirements become complicated, the intuition based designs may not work effectively as expected. Topology optimization, on the other hand, has the potential to systematically generate designs at their optimal performance without requiring a priori knowledge. This powerful tool has been utilized for decades and has seen a wide application in many engineering fields. Here in this thesis, the shape and topology optimization of designs are carried out within the framework of the level set approach. During the optimization, the design is implicitly represented by the zero level of a one-dimension-higher hypersurface which is referred as the level set function. By coupling the level set model with the physical model, the design performance can be evaluated in each optimization iteration to provide feedback to the designing process. With the feedback information, the level set function is updated accordingly until the final optimal structural layout is achieved. With the help of the level set methods, the topological change of the design, including splitting and merging can be handled in a natural way. | 164 pages

Download

Share

COinS