|
[Sponsors] |
April 25, 2012, 12:41 |
How to start learning CFD?
|
#1 |
New Member
Nithin Poduval
Join Date: Apr 2012
Location: India
Posts: 1
Rep Power: 0 |
Hii , this question might seem real dumb, but its because I'm quite new to CFD.
I want to gain a level of expertise so that I can write my own codes. How can I do this? i mean what all books to read, and more importantly what all topics I should know well. Should I start with the Finite difference method? |
|
April 25, 2012, 19:41 |
|
#2 |
Senior Member
Chris DeGroot
Join Date: Nov 2011
Location: Canada
Posts: 414
Rep Power: 17 |
If you are a student you should see if your university offers a course. What books you should read depends on which method(s) you want to learn about. Patankar's "Numerical heat transfer and fluid flow" is the standard first reference for finite volume methods. Ferziger and Peric's "Computational Methods for Fluid Dynamics" is also good. Writing a simple 2D structured finite volume code seems like a reasonable place to start.
|
|
April 26, 2012, 01:32 |
|
#3 |
New Member
Dr. Madhukar M Rao
Join Date: Nov 2010
Location: Bangalore, India
Posts: 21
Rep Power: 15 |
I would also recommend, "An introduction to computational fluid dynamics" by Versteeg and Malalasekeara. If you are interested in high Mach number flows, then "Computational Gasdynamics" by Laney is also a good place to start. Other introductory books: "Using Computational Fluid Dynamics" by C. T. Shaw, "Computational Fluid Dynamics" by John D. Anderson.
|
|
April 26, 2012, 13:36 |
|
#4 |
New Member
JMC
Join Date: Apr 2012
Posts: 9
Rep Power: 14 |
Do you know how to write code? If not then start with a book or course in programming. Matlab is a good start, but eventually you will want to know C, Fortran, etc. An introductory course in programming will cover solving PDE's I would suggest starting there.
|
|
April 26, 2012, 19:29 |
|
#5 |
New Member
Join Date: Apr 2012
Posts: 27
Rep Power: 14 |
How difficult could it be if you already know good how to program with a couple of programming languages?
C/C++ Python Visual Basic |
|
April 26, 2012, 20:37 |
|
#6 |
Member
Mosi Owa
Join Date: Nov 2011
Posts: 35
Rep Power: 14 |
knowing a programming language is something, writing a code is something else. However, being familiar with a programming language is inevitable to start writing a code. I believe that writing a good code (even a very basic one) needs both experience and good understanding of CFD concepts. The former is of higher importance which I personally lack.
|
|
April 26, 2012, 21:54 |
|
#7 |
New Member
JMC
Join Date: Apr 2012
Posts: 9
Rep Power: 14 |
It depends how well you understand the physics and numerical analysis. Learning CFD is a life-long process. The more that you use it, the more you will learn.
|
|
April 27, 2012, 03:29 |
|
#8 | ||
New Member
Join Date: Apr 2012
Posts: 27
Rep Power: 14 |
Quote:
Quote:
Exactly. I have programmed for years with c/c++ but I have to refresh FD, and numeric analysis is something new for me. I have a lot of work. |
|||
May 10, 2012, 23:20 |
|
#9 |
New Member
Patrick
Join Date: Apr 2012
Posts: 28
Rep Power: 14 |
You need to read a book on numerical methods if you don't know about that, a book on aerodynamics if you don't know anything about that, and I'd suggest Andersons book on CFD, it gives a good intro to CFD and also guides you through making your own code for 4 different simple simulations (1-D Nozzle flow, Couette Flow, and supersonic flow over a wing, off the top of my head)
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
What is the best resource (book) to start learning about CFD? | mahill | Main CFD Forum | 4 | May 6, 2011 17:38 |
CFD learning material | polymer | Main CFD Forum | 1 | April 25, 2011 06:51 |
How to start unsteady propeller CFD | Joe | Main CFD Forum | 1 | February 14, 2009 01:13 |
asking for books to start learning cfd | Prateek kumar Jain | Main CFD Forum | 3 | January 24, 2008 13:11 |
ASME CFD Symposium, Atlanta, 22-26 July 2001 | Chris R. Kleijn | Main CFD Forum | 0 | August 1, 2000 10:07 |