j****x
发帖数: 943 | 1 初学CFD, 正在研究 Free Surface Flow, 哪位高人给介绍一下有关 Arbitrary
Lagrangian-Eulerian 和 Level-Set 方面的资料? 谢了先!! |
w*********h
发帖数: 664 | 2 Documentation of FULENT software.
【在 j****x 的大作中提到】
: 初学CFD, 正在研究 Free Surface Flow, 哪位高人给介绍一下有关 Arbitrary
: Lagrangian-Eulerian 和 Level-Set 方面的资料? 谢了先!!
|
j****x
发帖数: 943 | 3 得自己写和改前人的code，所以想懂得具体全面点。Anyone？ |
m*****t
发帖数: 2800 | 4 感觉level set是比较新、比较学术的东西，才２０来年历史吧。 |
j****x
发帖数: 943 | 5 好像是比较新，Berkeley 的 Sethian 好像是最早研究的人之一。 不过，大多数主流
的商业软件好象都有了这个方法。对free surface flow 是一个很好的工具。 所以想
学学， 但感觉其中的数学还是挺难的，看paper无济于事。 而且对ALE也比较迷芒，所
以想问问有没有好的参考资料。
【在 m*****t 的大作中提到】
: 感觉level set是比较新、比较学术的东西，才２０来年历史吧。
|
l**h
发帖数: 112 | 6 ALE is an "old" method for moving surface and fluid-structure
interaction modeling. In solid mechanics, it is also used to
avoid numerical problems resulting from large deformation.
Use Google?
http://www.wiley.co.uk/ecm/pdfs/Volume_1_Chapter_14.pdf |
j****x
发帖数: 943 | 7
Read Chapter.14 before, pretty concise. I am not able to understand to the
level that I can implement it, probably due to my lack of math and
programming skills. Anyway, thanks a lot for the link.
Although ALE is "old", I was told that it is better and more precise when
applying interfacial stress conditions than level set or VOF methods. Am I
right?
【在 l**h 的大作中提到】
: ALE is an "old" method for moving surface and fluid-structure
: interaction modeling. In solid mechanics, it is also used to
: avoid numerical problems resulting from large deformation.
: Use Google?
: http://www.wiley.co.uk/ecm/pdfs/Volume_1_Chapter_14.pdf
|
k*********g
发帖数: 791 | |
k*********g
发帖数: 791 | 9 asking for level set & ale???
all these above comments about them are inaccurate; in particular, the ALE
in most literature is mathematically WRONG and the concept or vision about
ALE is WRONG;
my suggestion for you:
before moving onto fs, you should first complete two phases:
a. be able to derive all those weighted integral forms of mathematical
equations smoothly, with zero errors;
b. do a programming (solivng some fixed-geometry flow, in c++), from the
very first step to the very final step, |
k*********g
发帖数: 791 | 10 extra:
1. when derive weighted integral form of equations, make sure you must be
able to use tensor handily!!!
2. do remember: get rid of fortran, ASAP!!! |
|
|
j****x
发帖数: 943 | 11 First of all, thanks a lot for the valuable info.
Well, I am scared to know that "the ALE in most literature is mathematically
WRONG and the concept or vision about ALE is WRONG". Would you please post
one or two currently available mathematically correct paper? I suppose the
original Hirt ALE paper should be correct.
As far as the weighted integral form goes, I actually have somewhat
confidence. I took plenty of numerical analysis, FEM and BEM courses before,
programmed codes for simple and sli |
l**h
发帖数: 112 | 12 Can you give details on your comments?
"all these above comments about them are inaccurate; in particular, the ALE
in most literature is mathematically WRONG and the concept or vision about
ALE is WRONG;"
ALE is a relatively mature method, which has been used in both fluid and
solid mechanics for a long time. i don't know how people can get it wrong. |
l**h
发帖数: 112 | 13 i would suggest jerryx just start programming in fortran,
based on literature you have now.
if you wait until you fully understand everything and learn to use c++, it
will be months later.
programming actually helps you to understand the theory.
just a thought... |
j****x
发帖数: 943 | 14 thanks lywh. I am programming in Fortran now. well, step by step though and
I totally agree with you that programming helps to understand the theory. |
t***l
发帖数: 533 | 15 free surface flow主要有level set method 和VOF 方法
Level set最初是Osher & Sethian 1988年JCP上的文章
之后分别衍生了如fast matching, ghost fluid等等
Osher和Sethian分别出过LS的书，都是经典，
sethian的书适合入门，
【在 j****x 的大作中提到】
: 初学CFD, 正在研究 Free Surface Flow, 哪位高人给介绍一下有关 Arbitrary
: Lagrangian-Eulerian 和 Level-Set 方面的资料? 谢了先!!
|
k*********g
发帖数: 791 | 16 你妈妈有没有从小教你养成讲卫生的好习惯？
有不良嗜好是很难该的；
get rid of fortran；do c＋＋ now！！！
藕就是因为从5岁开始染上了下棋的恶习；
要不然，藕早就把numericals搞完了；
于是dr tu也不会死了； |
k*********g
发帖数: 791 | 17 these categorization is logically messy;
also, ghost fluid has nothing to do with level set; it was developed
independently; and it even should not be considered as a method; it is just
a little trick; if this is a method, then there are thousandsof methods,
then we are living in total confusion;
we shall name ibm, immersed boundary element, as a method; so that a bunch
of stuff could be put into the same set;
【在 t***l 的大作中提到】
: free surface flow主要有level set method 和VOF 方法
: Level set最初是Osher & Sethian 1988年JCP上的文章
: 之后分别衍生了如fast matching, ghost fluid等等
: Osher和Sethian分别出过LS的书，都是经典，
: sethian的书适合入门，
|
k*********g
发帖数: 791 | 18 你妈妈有没有从小教你养成讲卫生的好习惯？
有不良嗜好是很难该的；
get rid of fortran；do c＋＋ now！！！
藕就是因为从5岁开始染上了下棋的恶习；
要不然，藕早就把numericals搞完了；
于是dr tu也不会死了； |
k*********g
发帖数: 791 | 19 ibm, immersed boundary method;
sorry for typo; |
