Development of novel computational methods for optimal design of electrically small electromagnetic scattering particles
| Author(s) | Jensen, Sarah | |
| Date Accessioned | 2023-03-31T11:59:27Z | |
| Date Available | 2023-03-31T11:59:27Z | |
| Publication Date | 2022 | |
| SWORD Update | 2023-02-14T17:10:00Z | |
| Abstract | Metamaterials and metasurfaces are well-studied in the scientific community. They allow for unusual material properties over a wide range of engineering applications that would not otherwise be achievable. One interesting application is electromagnetic scattering from small particles. Past manufacturing capabilities limited interest in this study. However, current manufacturing technologies allow the creation of small particles with complex metallic patterns. Finding patterns that can minimize or maximize scattering is of general interest. Unfortunately, existing general-purpose commercial software packages are not well suited to finding optimized pattern designs that have any complexity due to the long computation times. A custom computational electromagnetic algorithm developed for this purpose is thus needed. ☐ In this dissertation, I present a computationally efficient custom electromagnetic solver used to find optimized patterns that either enhance or reduce scattering at a desirable frequency range. My approach, which is based on the standard method of moments (MoM) algorithm, allows for a computationally efficient rigorous solution from planar flakes with complex printed metallic patterns. This computationally efficient method was integrated within well-known iterative optimization algorithms (genetic algorithm, particle swarm, pattern search, etc.) to arrive at patterns with improved scattering properties. This technique was applied to three cases 1) maximizing the backscattering through the co-polarized radar cross section (RCS), 2) minimizing the forward scattering through the co-polarized C_ext, and 3) maximizing the backscattering through co- and cross-polarized RCS. For each case electrically small particles as small as 0.2 wavelengths on a side were found that significantly enhanced scattering compared to metal flakes of the same size. The algorithms were generalized to include material effects such as finite conductivity metals to evaluate potential performance degradation and glide symmetry to arrive at patterns that are less sensitive to registration errors that occur during fabrication. A particle cloud model was also developed to predict scattering from a cloud of randomly oriented particles. Lastly, one of the most promising particle designs was fabricated and tested in a custom make particle cloud chamber providing experimental validation. | |
| Advisor | Mirotznik, Mark S. | |
| Degree | Ph.D. | |
| Department | University of Delaware, Department of Electrical and Computer Engineering | |
| DOI | https://doi.org/10.58088/tdw1-ns55 | |
| Unique Identifier | 1374316306 | |
| URL | https://udspace.udel.edu/handle/19716/32608 | |
| Language | en | |
| Publisher | University of Delaware | |
| URI | https://login.udel.idm.oclc.org/login?url=https://www.proquest.com/dissertations-theses/development-novel-computational-methods-optimal/docview/2777425828/se-2?accountid=10457 | |
| Keywords | Antenna shape synthesis | |
| Keywords | Electrically small particle | |
| Keywords | Metasurfaces | |
| Keywords | Optimized scattering | |
| Title | Development of novel computational methods for optimal design of electrically small electromagnetic scattering particles | |
| Type | Thesis |