• Libraries
    • Log in to:
    View Item 
    •   MSpace Home
    • Faculty of Graduate Studies (Electronic Theses and Practica)
    • FGS - Electronic Theses and Practica
    • View Item
    •   MSpace Home
    • Faculty of Graduate Studies (Electronic Theses and Practica)
    • FGS - Electronic Theses and Practica
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Electromagnetic scattering by numerical methods applicable for large structures

    Thumbnail
    View/Open
    NQ53081.pdf (7.495Mb)
    Date
    2000-07-01
    Author
    Ye, Qiubo
    Metadata
    Show full item record
    Abstract
    The objective of this research is to develop numerical methods for general and efficient solutions to the linear systems obtained using the integral equations arising from electromagnetic scattering problems involving electrically large structures. In the process, the prior art in this area is reviewed. Then, the integral equations and their solutions by the method of moments (MoM) are derived. The progressive numerical method (PNM) and the projection iterative method (PIM) are analysed, including formulations, operation counts, stopping criteria, and their connection. In practice, the PNM is successful in calculation of two-dimensional scattering problems. The iterative PNM and a special case of the PNM, the modified spatial decomposition technique (SDT), are applied to the problems and compared with the PNM. Examples show that the PNM can depress internal resonances. The PIM is implemented in the two-dimensional TE case and convergent solutions are obtained. In order to overcome the difficulties with three-dimensional scattering problems, the PIM is implemented to solve the matrix equation obtained by MoM. Convergent results are observed in all examples being calculated for two- and three-dimensional objects. The PIM's iteration process can be accelerated by appropriate relaxation factors. The dependence of optimum relaxation factors on various parameters are investigated. Approximate results of large objects are obtained by the PIM with much less computation effort than the direct method. By allowing certain smaller elements in a coefficient matrix to be zero, the PIM can be further sped up, while still getting good far field results. This technique was found to be object dependent, providing better results for spheres than other objects.
    URI
    http://hdl.handle.net/1993/1882
    Collections
    • FGS - Electronic Theses and Practica [25494]

    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of MSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

    My Account

    Login

    Statistics

    View Usage Statistics

    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV