{"title":"基于完全重叠域分解的大型复合介质中小型天线的快速优化","authors":"T. Peng, K. Sertel, J. Volakis","doi":"10.1109/CEM.2009.5228103","DOIUrl":null,"url":null,"abstract":"A novel domain decomposition approach for modeling electrically small details within large computational domains is proposed. The approach tackles the conditioning issues associated with highly non-uniform finite element meshes resulting from direct discretization of fine geometrical details. The proposed method addresses this particular challenge, specifically for fast optimization tools, by separating the discretization of the detail region from that of the uniform background. The iterations between the two are carried out by proper field projections between the meshes. Due to two overlapping meshes being completely unstructured, the method avoids frequent re-meshing and matrix re-computation in optimization iterations. This aspect provides significant flexibility, compared to conventional finite element and domain decomposition methods.","PeriodicalId":416029,"journal":{"name":"2009 Computational Electromagnetics International Workshop","volume":"3 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":"{\"title\":\"Fully overlapping domain decomposition for fast optimization of small antennas in large-scale composite media\",\"authors\":\"T. Peng, K. Sertel, J. Volakis\",\"doi\":\"10.1109/CEM.2009.5228103\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A novel domain decomposition approach for modeling electrically small details within large computational domains is proposed. The approach tackles the conditioning issues associated with highly non-uniform finite element meshes resulting from direct discretization of fine geometrical details. The proposed method addresses this particular challenge, specifically for fast optimization tools, by separating the discretization of the detail region from that of the uniform background. The iterations between the two are carried out by proper field projections between the meshes. Due to two overlapping meshes being completely unstructured, the method avoids frequent re-meshing and matrix re-computation in optimization iterations. This aspect provides significant flexibility, compared to conventional finite element and domain decomposition methods.\",\"PeriodicalId\":416029,\"journal\":{\"name\":\"2009 Computational Electromagnetics International Workshop\",\"volume\":\"3 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2009-07-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"8\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2009 Computational Electromagnetics International Workshop\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CEM.2009.5228103\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 Computational Electromagnetics International Workshop","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CEM.2009.5228103","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Fully overlapping domain decomposition for fast optimization of small antennas in large-scale composite media
A novel domain decomposition approach for modeling electrically small details within large computational domains is proposed. The approach tackles the conditioning issues associated with highly non-uniform finite element meshes resulting from direct discretization of fine geometrical details. The proposed method addresses this particular challenge, specifically for fast optimization tools, by separating the discretization of the detail region from that of the uniform background. The iterations between the two are carried out by proper field projections between the meshes. Due to two overlapping meshes being completely unstructured, the method avoids frequent re-meshing and matrix re-computation in optimization iterations. This aspect provides significant flexibility, compared to conventional finite element and domain decomposition methods.