A Memory-Efficient PITD Method for Multiscale Electromagnetic Simulations

0 ENGINEERING, ELECTRICAL & ELECTRONIC IEEE microwave and wireless technology letters Pub Date : 2024-06-13 DOI:10.1109/LMWT.2024.3408457
Jiawei Wang;Minyu Mao;Ru Xiang;Huifu Wang;Haoyu Lian
{"title":"A Memory-Efficient PITD Method for Multiscale Electromagnetic Simulations","authors":"Jiawei Wang;Minyu Mao;Ru Xiang;Huifu Wang;Haoyu Lian","doi":"10.1109/LMWT.2024.3408457","DOIUrl":null,"url":null,"abstract":"A memory-efficient variant of the precise-integration time-domain (PITD) method is proposed for multiscale electromagnetic simulations involving geometry details in only one or two dimensions. In the classic PITD method, the dense matrix exponential of the time-stepping operator arising from the finite difference discretization needs explicit evaluation and storage, leading to prohibitive memory costs. In the proposed method, the precise integration (PI) method is used to efficiently compute the sparse matrix exponential of a diagonal operator to obtain a transformation of the original ordinary differential equation (ODE) system, which has a relaxed stability criterion and can be integrated by any explicit time integration scheme. It is demonstrated by numerical experiments that the proposed method can exclude the stiffness due to directional geometry details and outperforms the classic finite-difference time-domain (FDTD) method in multiscale analysis.","PeriodicalId":73297,"journal":{"name":"IEEE microwave and wireless technology letters","volume":"34 8","pages":"967-970"},"PeriodicalIF":0.0000,"publicationDate":"2024-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE microwave and wireless technology letters","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10555138/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"0","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

A memory-efficient variant of the precise-integration time-domain (PITD) method is proposed for multiscale electromagnetic simulations involving geometry details in only one or two dimensions. In the classic PITD method, the dense matrix exponential of the time-stepping operator arising from the finite difference discretization needs explicit evaluation and storage, leading to prohibitive memory costs. In the proposed method, the precise integration (PI) method is used to efficiently compute the sparse matrix exponential of a diagonal operator to obtain a transformation of the original ordinary differential equation (ODE) system, which has a relaxed stability criterion and can be integrated by any explicit time integration scheme. It is demonstrated by numerical experiments that the proposed method can exclude the stiffness due to directional geometry details and outperforms the classic finite-difference time-domain (FDTD) method in multiscale analysis.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
用于多尺度电磁模拟的内存效率 PITD 方法
针对只涉及一维或两维几何细节的多尺度电磁仿真,提出了精确积分时域(PITD)方法的内存高效变体。在经典的 PITD 方法中,有限差分离散化产生的时间步进算子的密集矩阵指数需要显式评估和存储,从而导致过高的内存成本。在所提出的方法中,精确积分(PI)方法用于高效计算对角线算子的稀疏矩阵指数,从而获得原始常微分方程(ODE)系统的变换,该变换具有宽松的稳定性准则,可通过任何显式时间积分方案进行积分。数值实验证明,所提出的方法可以排除方向性几何细节导致的刚度,在多尺度分析中优于经典的有限差分时域(FDTD)方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
CiteScore
6.00
自引率
0.00%
发文量
0
期刊最新文献
Table of Contents IEEE Open Access Publishing IEEE Microwave and Wireless Technology Letters publication IEEE Microwave and Wireless Technology Letters Information for Authors TechRxiv: Share Your Preprint Research with the World
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1