A Wave Superposition Method Formulated in Digital Acoustic Space
Open Access
- Author:
- Hwang, Yong Sin
- Graduate Program:
- Acoustics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- April 08, 2009
- Committee Members:
- Gary Hugo Koopmann, Dissertation Advisor/Co-Advisor
Gary Hugo Koopmann, Committee Chair/Co-Chair
Victor Ward Sparrow, Committee Chair/Co-Chair
Martin Wesley Trethewey, Committee Member
John Brian Fahnline, Committee Member - Keywords:
- Boundary Element Method
BEM
Voxel
Wave Superposition Method
Shape Optimization
Digital acoustic space - Abstract:
- In this thesis, a new formulation of the Wave Superposition method is proposed wherein the conventional mesh approach is replaced by a simple 3-D digital work space that easily accommodates shape optimization for minimizing or maximizing radiation efficiency. As sound quality is in demand in almost all product designs and also because of fierce competition between product manufacturers, faster and accurate computational method for shape optimization is always desired. Because the conventional Wave Superposition method relies solely on mesh geometry, it cannot accommodate fast shape changes in the design stage of a consumer product or machinery, where many iterations of shape changes are required. Since the use of a mesh hinders easy shape changes, a new approach for representing geometry is introduced by constructing a uniform lattice in a 3-D digital work space. A voxel (a portmanteau, a new word made from combining the sound and meaning, of the words, volumetric and pixel) is essentially a volume element defined by the uniform lattice, and does not require separate connectivity information as a mesh element does. In the presented method, geometry is represented with voxels that can easily adapt to shape changes, therefore it is more suitable for shape optimization. The new method was validated by computing radiated sound power of structures of simple and complex geometries and complex mode shapes. It was shown that matching volume velocity is a key component to an accurate analysis. A sensitivity study showed that it required at least 6 elements per acoustic wavelength, and a complexity study showed a minimal reduction in computational time.