[jchawner]

Daniel: I don't have a PhD. My degree is from the school of hard knocks.

[ibluesun]

Originally Posted by jchawner

Hello Ahmed:

You are to be congratulated for wanting to work on mesh generation. Your professor is correct - it is the hardest part of CFD and therefore the part where advancements are needed.

I'm not certain the world needs yet another grid generation software program. Instead, focus on some of the core issues that make mesh generation challenging. Develop methods, don't write a program. Several other people have already mentioned these things but I'll repeat some of them here.

Dealing with CAD geometry (defeaturing, healing, abstracting). Quantification of a priori mesh quality metrics on CFD solution convergence and accuracy. Unstructured hex meshing. Anisotropic unstructured meshing (i.e. stretching along flow features such as boundary layers, wakes, shock waves). Automatic decomposition of a 3D domain into zones (for structured grid blocking, for example). Parallel tet meshing. These are just a few ideas.

Best of luck with your research.

Hello John

Originally Posted by jchawner

Develop methods, don't write a program.

I spent a reasonable time in thinking about this sentence, somehow I claim that I started to understand what you are talking about ..

I appreciate your chosen words.

[ibluesun]

Originally Posted by lakeat

Recently, I have just written an amateur's review on current mesh generations. I think you make this too hard for ibluesun, he just started it.

My two cents is:
Since we dont know exactly what ibluesun's background is, so the way he focuses on mesh generation might be different than others. You all shared about your concern on CAD geometry import and export and healing issue, but is this so urgent for a beginner? For me, doing turbulence studies on some simple geometries, I dont care CAD issue at all.

So I agree with Dr. Chawner's doubt, --"do we need another mesh generation software", my point is this,
I feel some issues like what Dr. Chawner has mentioned: 1) "Automatic decomposition of a 3D domain into zones"; 2) "Quantification of a priori mesh quality metrics on CFD solution convergence and accuracy"; are actually more urgent than the others for a beginner, so that his contribution to mesh generation will not turn out to be solely a GUI, but a very cool mesh solver behind it. And the better he understand a mesh solver, and the better he will have a good philosophy for a mesh GUI design.


Last points, I think Mr. ibluesun will not and can not handle all the issues in limited years, so why not pick up one or two issue and delve into it.

Personally, I still think multi-block structured mesh is more valuable than unstructured mesh, they should be sought whenever possible.

my background in CFD is pathetic but I am trying to gain knowledge as much as I can

the idea is because of my programming skills my professor found it a chance to let me make this program, actually his requirements is more simple than the vision here, but I couldn't just sit without knowing the wider vision.

although I find it strange, but would be interresting if you checked this open source project http://SymbolicAlgebra.codeplex.com and also its usage inside http://quantitysystem.codeplex.com

the two are mine with the same ibluesun nick name also

those two projects will be my swiss knife in making my program.

and by working in my mesh generator i should be able to understand great deal on geometry (I hope so)


Edit: I am a mechanical engineer graduate

[ibluesun]

Originally Posted by cfdnewbie

Dear John,
What I mean by high order grids is the representation of curved geometry by elements which are high order accurate, e.g. high order polynomial mappings (let's say p>=4) with well-defined normal vectors. To my knowledge, the generation of curved high order meshes is an open issue, since most grid generators just produce linear elements..... Or has that been solved? I'm not completely up to date with the new features of grid generators, I admit!

hello cfdnewbie

thank you for you reply, actually it would be nice if you explained this issue for me, or pointed to a link that explain it

[ibluesun]

Originally Posted by Martin Hegedus

Hi Ahmed,

Here is my suggestion on an approach:
1) Talk to various people to get an idea of types of physics which create the biggest challenges for CFD.
2) Somehow get an idea of what CFD methodologies are required to capture these physics. This can include RANS, DNS, LES, and discretization methods, just to name a few. You don't necessarily need to understand the theory behind it, yet, but you do need to know what is out there.
3) Ask your professor if he/she can recommend some people in industry who use CFD that you can talk to one-on-one. Get an idea of how they use CFD, what their expectations are, and the direction they would like CFD to go.
4) Go back to your professor and have an in depth CFD discussion.
5) Factor 1-4 into your choice of paths forward in regard to gridding.

It is my opinion that there are two types of CFD out there, qualitative (get a basic idea of the flow) solutions and quantitative (nail down the numbers) solutions.

Getting qualitative results seems to be a very well populated market segment.

However, who and what is in the market segment for quantitative results is much more ambiguous. I think some like to give the impression that their qualitative results are quantitative, when they are not. Unfortunately, many times I can't sort out the truth from the fiction. It is also hard to determine how "good" or "bad" a result is.

Getting quantitative results is a real pain, at least for me. And it is not just the gridding. It is the whole process.

Take something as simple as a vortex. Tracking it properly can be a challenge, especially when it is interacting with flow features and geometries. And tracking it properly requires a bit of gridding and CFD solver work. The unfortunate part is that the vortex trajectory and strength is sensitive to the grid, time step, and the order of the spatial discretization. How "good" higher order spatial discretization is depends on how "good" the grid is. Higher order discretization tends to be sensitive to the grid.

So now we get, in my opinion, to a basic question. For quantitative results where one needs to resolve severe nonlinearities, (shocks, wake on geometry interactions, tracking a vortex, etc) is a structured or unstructured grid better? I don't have the answer to this. And the choice of structured vs. unstructured is not just based on geometry, gridding, and whether the CFD formulation is implicit or explicit, it's also based on computer hardware. In general, structured grids are more efficient with the memory bus.

You can find examples of structured and unstructured grid results at the drag prediction workshop, http://aaac.larc.nasa.gov/tsab/cfdlarc/aiaa-dpw/ (at the bottom of the page you'll see links to the older workshops) And don't forget, these are nice clean configurations. Sure, they have some physics excitement, but not nearly as much as they could! (Personally, I think it would be more fun to do the computations of one aircraft in the near wake of another!)

Edit: My example of structured vs. unstructured is not necessarily connected with Daniel's comment since I posted my reply before I read his post.

Hi Martin

your approach is definitely right.
so quantitive result should be my orientation.

my professor showed how was the solution was bad if I made a grid that is opposite to the flow (beside telling me the difference between conserved and nonconserved governing equations)

here you are adding something new to my mind which is the type of flow itself (am I right here??) so type of flow is not its direction only but its nature may be (like the wake, and vortex example)

about RANS, LES (I know that they are modeling for the turbulent flow) I know also that turbulent flow equations has this strange behavior of fluctuated properties (and as I remember internal momentum exchange)

how could that affect my gridding (i'll leave it to the days to tell me )

very much appreciated for widen my vision more

[ibluesun]

your invaluable replies let me thought about this

do we have a CFD terms dictionary that makes a definition for all of these terms ??

and when I mean dictionary I mean a dictionary that has an approved definitions from experienced cfd users, and researchers.

this is just a question that came out from my head.

and really really thank you for your invaluable notes

[lakeat]

Are you looking for CFD jargon?

Another better way is to ask for John or Martin's Skype, they are walking dictionaries. haha

[ibluesun]

searched and found this PDF ,
How to –
Understand Computational Fluid Dynamics Jargon
Editing Author:
Althea de Souza 2005 , i think yes this is the answer for my question

[Martin Hegedus]

Originally Posted by ibluesun

here you are adding something new to my mind which is the type of flow itself (am I right here??) so type of flow is not its direction only but its nature may be (like the wake, and vortex example)

That is correct.

about RANS, LES (I know that they are modeling for the turbulent flow) I know also that turbulent flow equations has this strange behavior of fluctuated properties (and as I remember internal momentum exchange)

how could that affect my gridding (i'll leave it to the days to tell me )

The simple, and not necessarily complete nor 100% accurate all the time, answer is that the grid itself adds noise into the solution because of discretization and approximations. As the flow (actually velocity and pressure waves) goes through these cells it accumulates these errors, or more appropriately noise. Low order spatial descritization methods will have either artificial dissipation or a good dose of internal dissipation (time marching compressible codes). This dissipation either covers the noise or smooths/eliminates it. However, as the order of the spatial descritization increases the need for artificial dissipation decreases or the amount of internal dissipation decreases, and the grid noise starts to come to the forefront. If the flow goes through one noisy cell, probably not an issue. If the flow goes through many noisy cells, its a bigger issue. A RANS code, in general, is not that sensitive to the noise. The eddy viscosity covers/smooths out the noise. Though, if there is enough noise a steady solution can be kicked into being unsteady or the accuracy of an unsteady solution can be affected. On the other hand, a LES solution, since it is trying to resolve smaller flow features which turn into big flow features, is sensitive to the noise. It doesn't take much to give it a kick. Or so I understand. I don't claim to be an expert in LES. To thoroughly play around with LES requires a lot of CPU power.

[cfdnewbie]

Originally Posted by ibluesun

hello cfdnewbie

thank you for you reply, actually it would be nice if you explained this issue for me, or pointed to a link that explain it

Hello ibluesun,
here's a link that might shed some light:
http://www.simtech.uni-stuttgart.de/...nts.php?ID=156

[Eezyville]

I recently went to a workshop on mesh generation. It was sponsored by the Air Force and held at WPAFB. They had some big names there like Pointwise, Hypermesh, and Fluent. The biggest problem that they talked abut was getting good quality meshes. They showed examples of how skewedness effected meshes at different flight regimes. I can give you some more details on it but if you want to do something useful then look into what is needed to make high quality, highly accurate meshes. This may ultimately depend on the specific problem but everyone there had different solutions or no solutions whatsoever.

[kamakura117]

Personally, I think that a GUI is a nice way to go, but a GUI in the sense that it writes its own macros that it reads and run. This gives the user the capability to then script their own meshing. I really like Cd-Adapco's meshing capabilities in STAR-CCM+ v6-7. It automates much of the process in determining the geometry constraints and also allows for plenty of user optimization of the mesh. It also patches holes and fits the CAD model as needed. It is a commercial code, so there is no access to its routines. However, from a user-aspect, I think that would be a good guideline for emulation since it allows for scripting and provides a very intuitive interface, and it does produce computationally efficient/accurate meshes (in most cases, anyways).

I would also recommend a using a language that is completely cross-platform. Some might giggle: but Java is actually fairly robust and fast now-a-days. It would be a easy transition from C#, too.

[Al-Hashimi]

HI Ahmed, can you please advice me how to create the geometry of Hydrocycolne? can i use the Auto-CAD?

[ibluesun]

I am happy to say that I reached a reasonable state for my program .. to the extent I can show you a demo.

http://youtu.be/nvQDcTJabQI


I opened a new thread for it at
http://www.cfd-online.com/Forums/mai...-software.html

I still have a long way to finish it .. I hope that I didn't disappoint you with this demo

[ibluesun]

hello guys .. I've opened a new thread for the latest updates I made on my meshing program.

www.cfd-online.com/Forums/main/139078-fantastic-mesh-modeler.html

would really like to hear your opinions after all.

[ibluesun]

Hello guys,

just to keep you updated I've added the ability to export to OpenFOAM

https://www.youtube.com/watch?v=F3EqdVrityk

and made a facebook page for it.

https://www.facebook.com/pages/Fanta...67869366677159

really appreciate all of your comments that helped me to reach this state.

Cheers,

[mechiebud]

Hi Ahmed,
Since you have worked on mesh generation,I hope you could guide me in my problem.
I am working on hyperbolic grid around airfoils. I have done that part. But I ma not getting a smooth mesh at trailing edge. I want to smooth the trailing edg with the help of ellipic functions. Do you have any idea of that?

[jchawner]

Hello Mech:

Do you want to include elliptic smoothing within the hyperbolic formulation so it's applied as the grid is marching out? Or do you want to apply it after the fact?

If the former, I have no idea. Your best approach within the hyperbolic method is to properly implement all the smoothing controls.

If the later, see Thompson's book on structured grids which is freely available online.

[mechiebud]

Originally Posted by jchawner

Hello Mech:

Do you want to include elliptic smoothing within the hyperbolic formulation so it's applied as the grid is marching out? Or do you want to apply it after the fact?

If the former, I have no idea. Your best approach within the hyperbolic method is to properly implement all the smoothing controls.

If the later, see Thompson's book on structured grids which is freely available online.

Hello jchawner and thanks for your response

I have already done the hyperbolic part. I have generated hyperbolic mesh around airfoil but I am not getting a very smooth mesh at trailing edge. So what I want to do is generate an elliptic mesh just at few points on the trailing edge and leave the hyperbolic edge on the remaining parts. Though I am not sure if it will work or not?

[jchawner]

I just want to be certain of the terms you're using. The hyperbolic PDE method generates the points. An elliptic PDE method smooths existing points. So the way you describe it, you'd generate your mesh first with the hyperbolic PDE method and then apply an elliptic PDE method to a subset of that mesh in order to increase smoothness.