Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
dissertation_tosubmit.pdf (3.11 MB)
ETD Abstract Container
Abstract Header
Development of A Fast Converging Hybrid Method for Analyzing Three-Dimensional Doubly Periodic Structures
Author Info
Wang, Feng
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=osu1376923791
Abstract Details
Year and Degree
2013, Doctor of Philosophy, Ohio State University, Electrical and Computer Engineering.
Abstract
In this dissertation, we present an efficient hybrid method based on finite element (FE) and rigorous coupled wave analysis (RCWA) to model scattering from 3-D doubly periodic dielectric structures, which has been extensively studied over decades due to its wide applications. Double periodicity RCWA method, which reduces the diffraction problem to an eigenvalue problem, is mostly used for analyzing simple periodic structures. For arbitrary shape, it suffers from slow convergence and soon becomes impractical as its computational cost grows as M^6, where M is related to the Fourier harmonics included in analysis. The 3-D finite element method (FEM) is inherently suitable for modeling arbitrary shape, however, at the cost of increased number of unknowns. Fortunately, the proposed FE/RCWA method is a balanced approach, which provides better flexibility for modeling arbitrary structures over the double periodicity RCWA method, and significantly reduces the number of unknowns compared to 3-D FEM. To further reduce the size of the problem, the mode matching method (MMM) is chosen as boundary truncation scheme among other choices such as perfectly matched layer (PML), absorbing boundary condition (ABC) and boundary integral method (BIE). MMM not only allows the truncation boundaries to be placed very close to the periodic structures, but most importantly, has the property to match evanescent diffraction orders. For comparison, the employment of PML often requires large air regions between periodic structures and truncation boundaries to eliminate reflections. We also investigate the slow convergence behavior of the proposed FE/RCWA method under certain conditions. By incorporating fast Fourier factorization (FFF) method into the FE/RCWA formulation, we significantly improve the convergence rate for TE excitation. In addition, because the FFF method preserves the surface normal vectors of the original geometry, the gap between converged results under different polarizations is order of magnitude smaller than that of original FE/RCWA method. The advantage of FE/RCWA with FFF is demonstrated by a variety of examples.
Committee
Robert Lee (Advisor)
Fernando Teixeira (Committee Member)
Robert Burkholder (Committee Member)
Pages
160 p.
Subject Headings
Electrical Engineering
;
Electromagnetics
Keywords
periodic structures, finite element method, RCWA, mode matching, fast Fourier factorization
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Wang, F. (2013).
Development of A Fast Converging Hybrid Method for Analyzing Three-Dimensional Doubly Periodic Structures
[Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1376923791
APA Style (7th edition)
Wang, Feng.
Development of A Fast Converging Hybrid Method for Analyzing Three-Dimensional Doubly Periodic Structures.
2013. Ohio State University, Doctoral dissertation.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=osu1376923791.
MLA Style (8th edition)
Wang, Feng. "Development of A Fast Converging Hybrid Method for Analyzing Three-Dimensional Doubly Periodic Structures." Doctoral dissertation, Ohio State University, 2013. http://rave.ohiolink.edu/etdc/view?acc_num=osu1376923791
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
osu1376923791
Download Count:
1,020
Copyright Info
© 2013, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.