怎么学习计算电磁学【QUORA】
There are several resources. But it depends on what you actually want to learn...Let me explain:
First of all, There are many Numerical Modeling Techniques in Electromagnetic, and people are always confused about which one to choose. But if you are not able to decide, then best thing is to start with any of them (I am not talking about conventional theoretical approaches which cannot be done theoretically for bigger problem spaces). All numerical techniques can be used for all type of EM problems but major aspect is efficient solution, in terms of results and resources.
1. If you want to know about general idea about CEM, then its better you begin with Wikipedia itself, which has a great amount of details about numerical modeling in electromagnetics.
Computational electromagnetics
2. If you are not a beginner, then you might have already chosen, which numerical modeling you are interested in, like FDTD, FEM, MoM, FIT, etc.
If you have not chosen yet and want to decide which one to be considered for your numerical solutions of EM problems, then do not worry. Dr. Raymond Rumpf has created many lectures on CEM on youtube. So take a look at them.
If you want to go into full depth mathematical analysis then go through Dr. Cynthia Furse's videos on youtube.
I loved these two...I guess you may also like it at least. :)
3. If you have decided now, which numerical modeling you are interested in, then there are several books and videos available to learn each and every one of them.
There are many numerical modeling techniques, some of them you might have seen on Wikipedia itself.
But it depends on the EM problem, which decides choice the efficient way of numerical solution and numerical modeling method. So once you have decided it according to your EM problem then only start working on specific numerical modeling.
I like mainly following Numerical Techniques:
I). Finite Difference Time Domain Method (FDTD)
There are youtube videos to understand it easily but I would rather prefer if you go through:
Prof. Kane Yee's fundamental paper (1966). This paper itself is more than sufficient to understand FDTD for EM problems.
http://ecee.colorado.edu/~mcleod...
And a paper by Prof. Taflove
http://www.ece.northwestern.edu/...
Trust me if you can understand these, it is more than sufficient to begin with FDTD. I started with these only. Then there are many papers by Prof. Allen Taflove, Prof. Karl S. Kunz and many others.
First good book on FDTD by Kunz & Luebbers is like an epic for numerical scientists.
The Finite Difference Time Domain Method for Electromagnetics
Then there are many books by many professors, Prof. Taflove, Prof. Sadiku, Prof. Elsherbeni. You can just google them and search for free pdfs also. :) I will not put links here. :)
Even if you are interested in Videos only, then go through lecture series by Prof. C. Furse:
If you want to use commercially available Tools then:
i. FEKO -- https://www.feko.info/
ii. CST -- https://www.cst.com/
iii. FDTD Solutions -- Lumerical's Nanophotonic FDTD Simulation Software
Free FDTD codes and tools are
- FDTD++: advanced, fully featured FDTD software, with included C++ source code, along with sophisticated material models and predefined fits as well as discussion/support forums and email support
- openEMS (Fully 3D Cartesian & Cylindrical graded mesh EC-FDTD Solver, written in C++, using a Matlab/Octave-Interface)
- pFDTD (3D C++ FDTD codes developed by Se-Heon Kim)
- JFDTD (2D/3D C++ FDTD codes developed for nanophotonics by Jeffrey M. McMahon)
- WOLFSIM (NCSU) (2-D)
- Meep (MIT, 2D/3D/cylindrical parallel FDTD)
- (Geo-) Radar FDTD
- bigboy (unmaintained, no release files. must get source from cvs)
- toyFDTD
- Parallel (MPI&OpenMP) FDTD codes in C++ (developed by Zs. Szabó)
- FDTD code in Fortran 90
- FDTD code in C for 2D EM Wave simulation
- Angora (3D parallel FDTD software package, maintained by Ilker R. Capoglu)
- GSvit (3D FDTD solver with graphics card computing support, written in C, graphical user interface XSvit available)
- gprMax (Open Source (GPLv3), 3D/2D FDTD modelling code in Python/Cython developed for GPR but can be used for general EM modelling.)
These are taken from Wikipedia...
II). Method of Moments (MoM)
This is also one of the oldest methods in electromagnetics for solving big problems.
There is one good book by Gibson: Search on google and first link... :)
https://www.google.de/url?sa=t&r...
I have read this book and its awesome...you need patience ... :)
There are many books with MATLAB coding of MoM. If you want to implement it with MATLAB, then Go Through:
Fundamentals of Electromagnetics with MATLAB, By Karl Erik Lonngren, Sava Vasilev Savov, Randy
Commercial Tools:
i. FEKO --https://www.feko.info/ --------Best available software for MoM
ii. NEC codes --Numerical Electromagnetics Code NEC2 unofficial home page
III). Finite Element Method (FEM)
For 2D solutions, to begin with, just go through:
http://deepblue.lib.umich.edu/bi...
There are many books on FEM for Electromagnetics, but I would rather prefer it for specific numerical problems only. Or maybe inside FEKO with hybridizing with MoM.
Commercial Tools:
i. FEKO --https://www.feko.info/ --------Best available software for MoM
ii. HFSS -- http://www.ansys.com/Products/Si...
IV). Finite Integration Technique (FIT)
It was proposed by Prof. Thomas Weiland. IT has basic papers written by Prof. Weiland. There is a commercially available tool company, which is founded by Prof. Weiland and his colleagues only, known as CST Studio Suite.
CST -- https://www.cst.com/
------------------------------------------------------------------------------
There are many enhancements in these numerical modeling techniques and hybridization with many other numerical techniques (MLFMM, PO, GO etc.) for specific applications. For more information use the Wikipedia page which I have posted first... :)
-----
I hope this is useful to someone... :)
怎么学习计算电磁学【QUORA】的更多相关文章
- Caffe深度学习计算框架
Caffe | Deep Learning Framework是一个清晰而高效的深度学习框架,其作者是博士毕业于UC Berkeley的 Yangqing Jia,目前在Google工作.Caffe是 ...
- MXNET:深度学习计算-GPU
mxnet的设备管理 MXNet 使用 context 来指定用来存储和计算的设备,例如可以是 CPU 或者 GPU.默认情况下,MXNet 会将数据创建在主内存,然后利用 CPU 来计算.在 MXN ...
- MXNET:深度学习计算-自定义层
虽然 Gluon 提供了大量常用的层,但有时候我们依然希望自定义层.本节将介绍如何使用 NDArray 来自定义一个 Gluon 的层,从而以后可以被重复调用. 不含模型参数的自定义层 我们先介绍如何 ...
- MXNET:深度学习计算-模型参数
我们将深入讲解模型参数的访问和初始化,以及如何在多个层之间共享同一份参数. 之前我们一直在使用默认的初始函数,net.initialize(). from mxnet import init, nd ...
- MXNET:深度学习计算-模型构建
进入更深的层次:模型构造.参数访问.自定义层和使用 GPU. 模型构建 在多层感知机的实现中,我们首先构造 Sequential 实例,然后依次添加两个全连接层.其中第一层的输出大小为 256,即隐藏 ...
- python学习===计算字符串中子串出现的次数。
if __name__ == '__main__': str1 = input('input a string: \n') str2 = input('input a sub string: \n') ...
- 【深度学习Deep Learning】资料大全
最近在学深度学习相关的东西,在网上搜集到了一些不错的资料,现在汇总一下: Free Online Books by Yoshua Bengio, Ian Goodfellow and Aaron C ...
- 机器学习(Machine Learning)&深度学习(Deep Learning)资料(Chapter 2)
##机器学习(Machine Learning)&深度学习(Deep Learning)资料(Chapter 2)---#####注:机器学习资料[篇目一](https://github.co ...
- [Computational Advertising] 计算广告学笔记之基础概念
因为工作需要,最近一直在关注计算广告学的内容.作为一个新手,学习计算广告学还是建议先看一下刘鹏老师在师徒网的教程<计算广告学>. 有关刘鹏老师的个人介绍:刘鹏现任360商业产品首席架构师, ...
随机推荐
- docker与虚拟机性能比较
http://blog.csdn.net/cbl709/article/details/43955687www.chenbiaolong.com 概要 Docker是近年来新兴的虚拟化工具,它可以和虚 ...
- mysql启动不起来了!
[root@iZ28r2sl9qkZ data]# service mysqld restartMySQL server PID file could not be found! [FAILED]St ...
- (旧)子数涵数·Flash——影片剪辑的其他操作
一.复制影片剪辑 1.方法:duplicatemovieClip(影片实名,新实名,深度级别) 2.解释:影片实名就是你要复制的对象,新实名就是你要粘贴的对象,深度级别就是粘贴后的影片剪辑的堆叠顺序( ...
- ahjesus配置vsftpd和xinetd
vsftpd的简单配置参考此教程 传送门 教程内xinetd的配置失效, 用xinetd方式启动ftp方式如下 1.在/etc/xinetd.d/目录中创建一个文件vsftpd 内容如下: servi ...
- 关于使用ResultSet ---结果集没有当前行
Resultset返回的结果集是从结果的前一句开始的,也就是说一开始的resultset是没有的,所以,一般使用的时候,需要这样子写: while( resultSet.next() ){ ...
- swift学习笔记之-闭包
//闭包 import UIKit /*闭包(Closures): 函数.闭包.类都是引用类型(引用类型的实例赋值给变量或常量时,得到的都是该实例的引用,而值类型的实例变量得到的是独立的值的拷贝) 1 ...
- 流媒体一些server
(1)darwin stream server (2)red5 (3)nginx rtmp
- Swing(一):JFrame框架窗体
Swing窗体是一个组件,也是可视化的窗体,可以将其他组件放在这里.Jfream框架是一个容器,是Swing程序中各个组件的载体,可以将它看做为 一个容器,在开发中可以通过java.swing.jfr ...
- UISlider显示进度(并且实现图片缩放)
图片展示效果如下: 其他没什么好说的,直接上代码: RootView.h: #import <UIKit/UIKit.h> @interface RootView : UIView @pr ...
- iOS 验证邮箱手机号格式
做登录界面时,用户在UITextfield中输入输入邮箱账号后,我们应该在本地验证格式是否正确,再将参数传给服务器验证. 最简单的就是利用系统的NSPredicate //利用正则表达式验证 -(BO ...