Pub Date : 2023-01-17DOI: 10.1109/JMMCT.2023.3237699
Mengmeng Li;Yuchenxi Zhang;Zixuan Ma
Planar metasurfaces have been applied in several fields. Near-field coupling is typically neglected in traditional metasurface designs. A numerical modeling method for macrocells that considers near-field couplings between meta-atoms is proposed. A deep neural network (DNN) is constructed to accurately predict the electromagnetic response from different macrocells. Transfer learning is employed to reduce the number of the training datasets. The designed neural network is embedded in the optimization algorithm as an effective surrogate model. Both the deflector and high numerical aperture (NA) metalens are simulated and optimized with our design framework, approximately 30% improvements of efficiencies are achieved.
{"title":"Deep-Learning-Based Metasurface Design Method Considering Near-Field Couplings","authors":"Mengmeng Li;Yuchenxi Zhang;Zixuan Ma","doi":"10.1109/JMMCT.2023.3237699","DOIUrl":"https://doi.org/10.1109/JMMCT.2023.3237699","url":null,"abstract":"Planar metasurfaces have been applied in several fields. Near-field coupling is typically neglected in traditional metasurface designs. A numerical modeling method for macrocells that considers near-field couplings between meta-atoms is proposed. A deep neural network (DNN) is constructed to accurately predict the electromagnetic response from different macrocells. Transfer learning is employed to reduce the number of the training datasets. The designed neural network is embedded in the optimization algorithm as an effective surrogate model. Both the deflector and high numerical aperture (NA) metalens are simulated and optimized with our design framework, approximately 30% improvements of efficiencies are achieved.","PeriodicalId":52176,"journal":{"name":"IEEE Journal on Multiscale and Multiphysics Computational Techniques","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49962819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-13DOI: 10.1109/JMMCT.2023.3236946
Shutong Qi;Costas D. Sarris
This article presents the coupling of the finite-difference time-domain (FDTD) method for electromagnetic field simulation, with a physics-informed neural network based solver for the heat equation. To this end, we employ a physics-informed U-Net instead of a numerical method to solve the heat equation. This approach enables the solution of general multiphysics problems with a single-physics numerical solver coupled with a neural network, overcoming the questions of accuracy and efficiency that are associated with interfacing multiphysics equations. By embedding the heat equation and its boundary conditions in the U-Net, we implement an unsupervised training methodology, which does not require the generation of ground-truth data. We test the proposed method with general 2-D coupled electromagnetic-thermal problems, demonstrating its accuracy and efficiency compared to standard finite-difference based alternatives.
{"title":"Electromagnetic-Thermal Analysis With FDTD and Physics-Informed Neural Networks","authors":"Shutong Qi;Costas D. Sarris","doi":"10.1109/JMMCT.2023.3236946","DOIUrl":"https://doi.org/10.1109/JMMCT.2023.3236946","url":null,"abstract":"This article presents the coupling of the finite-difference time-domain (FDTD) method for electromagnetic field simulation, with a physics-informed neural network based solver for the heat equation. To this end, we employ a physics-informed U-Net instead of a numerical method to solve the heat equation. This approach enables the solution of general multiphysics problems with a single-physics numerical solver coupled with a neural network, overcoming the questions of accuracy and efficiency that are associated with interfacing multiphysics equations. By embedding the heat equation and its boundary conditions in the U-Net, we implement an unsupervised training methodology, which does not require the generation of ground-truth data. We test the proposed method with general 2-D coupled electromagnetic-thermal problems, demonstrating its accuracy and efficiency compared to standard finite-difference based alternatives.","PeriodicalId":52176,"journal":{"name":"IEEE Journal on Multiscale and Multiphysics Computational Techniques","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2023-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49962820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-12DOI: 10.1109/JMMCT.2023.3236645
Shuzhan Sun;Dan Jiao
In this work, we propose a new split-field domain-decomposition (DD) algorithm. Different from conventional DD methods where interface fields are treated as a whole and shared in common between adjacent subdomains, we split the field on the interface into $m$