CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   CFD of flapping wing mav (http://www.cfd-online.com/Forums/main/108901-cfd-flapping-wing-mav.html)

sultan November 4, 2012 11:41

CFD of flapping wing mav
 
i am trying to do fluid structure interaction of flapping wing mav as part of my final year project of bachelor of aerospace engineering,but before doing fsi i need to do cfd of flapping wing.so please guide me how to do cfd of flapping wing mav

cdegroot November 4, 2012 12:15

What software do you have access to? ANSYS can do FSI. If you have access look up the tutorials and if you have problems post to the appropriate sub forum with a specific question or problem

sultan November 4, 2012 13:50

i am to use ansys or nastran.which one would be better for cfd of flapping wing?

cdegroot November 4, 2012 15:47

I have no experience with Nastran, but Ansys can do FSI. It's a tough problem though. See what you can do with a simplified geometry. If you have problems these forums are a great source of help.

sultan November 5, 2012 04:03

Have you done it before?or has somebody else you know

cdegroot November 5, 2012 07:57

I haven't done a flapping wing nor do I know anyone that has. Do you need to do FSI or can it just be a moving mesh problem? i.e. do you care about the mechanical deformation of the wing and its interaction with the flow or can you assume it is a rigid wing moving in the flow. It's easier if it's just rigid. The hard part is that moving meshes can be less robust when the deformation is large.

sultan November 5, 2012 08:48

Actually i am to do one-way fsi of flapping wing my geometric model will only be wings of the micro air vehicle executing flapping motion.in one-way fsi there is no feedback loop i.e the aerodynamic loads wont change as a result of deformation(if deformations are considered small enough that they dont change the aerodynamics of the body then i think one way fsi is sufficient).can i consider the wing as rigid and run its cfd to find aerodynamic loads and then use those loads to find deformations?would doing cfd of rigid wing executing flapping motion like a bird be easy or doable?

cdegroot November 5, 2012 10:09

I don't want to make too many claims because I am not an expert in FSI; just an occasional user of such models in commercial software. I would say that if you are interested in the deformations due to the loads you are doing the right thing by choosing one-way FSI. I think this is easier than running the flow calculation followed by a calculation of the deformation.

I am hoping someone else can weigh in on your question. I know ANSYS software is capable of running the calculations you mention, I just don't have the experience to say whether it is doable or not.

If you want my opinion, I would not start with the full wing geometry you want to solve in the end. Start with something simpler. Maybe take a cylinder and make it move in the domain to see if you can get that to work. You want something that will run quickly to test the setup before you commit to the final geometry which will probably take a while to run. In CFX there are a few tutorial problems on two-way FSI. Do these first to get familiarized with the program.

sultan November 5, 2012 10:59

Thank you for your help im obliged.as i am new to this forum please guide me how can i attract other experts to answer this post.
Are u a student?which field?

cdegroot November 5, 2012 11:07

You are welcome for the help. For more help, you should post to the sub-forum for the software you are using (i.e. Fluent, CFX, etc.). You will find lots of help there. I read the CFX forum regularly. As for me, I have a PhD in Mechanical Engineering specializing in CFD and am currently working at a University in collaboration with one of the major CFD companies on some development work.

sultan November 6, 2012 09:12

Whats the difference between cfx and fluent?which one would be easier to use in case of my problem?

cdegroot November 6, 2012 09:55

The main difference between CFX and Fluent is that CFX is based on a control volume finite element method (also called vertex centred finite volume method) and Fluent is based on a cell centred finite volume method. Both programs should work for your problem. My preference is CFX because I find that it is more user friendly and I like the coupled solver (Fluent has a coupled solver too, but seems like more people use it's segregated solver). The thing I like best about CFX is the CFX Expression Language which makes it very easy to define custom boundary conditions, source terms, etc. Fluent has User Defined Functions, but they are not as user friendly in my opinion (they require coding and compiling). This is just a personal preference though; Fluent is also a good code.

daldry February 28, 2013 20:01

Hello,

I want to do the same thing! have you found some resources or tutorials!?


All times are GMT -4. The time now is 22:31.