Need help, Optimization Airfoil by C programming
I have read a search about optimization airfoil used Differential Evolution in C programming, and I want to know detail this code working. So, somebody has code or can write this code, please share! this is a directory of code:
Optimize parallel/
--File
--RUN_SIMULATION.sh
--airfoil
--air14
--airfoil_coordinates.dat
--params.txt
--mesh
--VAWT.glf
--post
--average_torque.dat
--avg.c
--avgmom
--mom
--run
--Generate_airfoil.sh
--Generate_mesh.sh
--Postprocess.sh
--Run solver.sh
--solver
--buil_case.jou
--flinput
--analysis
--dqs.sh
--dqs.sh.orig
--runme
--de
--in.dat
--params.txt
--runme
--screen.out
they were using airfoil NACA 0015 in this research and meshing by Ansys or Pointwise, torque calculates by Ansys.
Optimize parallel/
--File
--RUN_SIMULATION.sh
--airfoil
--air14
--airfoil_coordinates.dat
--params.txt
--mesh
--VAWT.glf
--post
--average_torque.dat
--avg.c
--avgmom
--mom
--run
--Generate_airfoil.sh
--Generate_mesh.sh
--Postprocess.sh
--Run solver.sh
--solver
--buil_case.jou
--flinput
--analysis
--dqs.sh
--dqs.sh.orig
--runme
--de
--in.dat
--params.txt
--runme
--screen.out
they were using airfoil NACA 0015 in this research and meshing by Ansys or Pointwise, torque calculates by Ansys.
Am I right in thinking you have just listed the files in a directory?!!!
By the look of it, somebody has been running ANSYS Fluent, probably in batch mode.
No idea what they are doing with it, but that VAWT.glf looks like "Vertical Axis Wind Turbine", yes? They are also after "torque" at one point, which fits. What exactly are they optimising? Blade pitch? Blade chord? Number of blades? NACA0015 is a single (symmetric) airfoil section.
If you want to run a 2-d or 3-d CFD simulation and get axial torque, then run Fluent:
https://www.ansys.com/en-gb/products/fluids/ansys-fluent
You can look in the CFD Online forum for Fluent at https://www.cfd-online.com/Forums/fluent/
If you are simply after aerofoil properties look at
http://airfoiltools.com/
Most of the basic drag and lift properties for these airfoils are derived with 2-d calculations using XFoil ( https://web.mit.edu/drela/Public/web/xfoil/ )
If you want to commission someone to write code for you then post in the Jobs Section.
If you simply want advice you will have to be a lot more explicit about what exactly you are doing.
EDIT: It's not this dratted reference is it?
https://www.hindawi.com/journals/isrn/2012/528418/
By the look of it, somebody has been running ANSYS Fluent, probably in batch mode.
No idea what they are doing with it, but that VAWT.glf looks like "Vertical Axis Wind Turbine", yes? They are also after "torque" at one point, which fits. What exactly are they optimising? Blade pitch? Blade chord? Number of blades? NACA0015 is a single (symmetric) airfoil section.
If you want to run a 2-d or 3-d CFD simulation and get axial torque, then run Fluent:
https://www.ansys.com/en-gb/products/fluids/ansys-fluent
You can look in the CFD Online forum for Fluent at https://www.cfd-online.com/Forums/fluent/
If you are simply after aerofoil properties look at
http://airfoiltools.com/
Most of the basic drag and lift properties for these airfoils are derived with 2-d calculations using XFoil ( https://web.mit.edu/drela/Public/web/xfoil/ )
If you want to commission someone to write code for you then post in the Jobs Section.
If you simply want advice you will have to be a lot more explicit about what exactly you are doing.
EDIT: It's not this dratted reference is it?
https://www.hindawi.com/journals/isrn/2012/528418/
Last edited on
Asking someone to post even a semi-serious CFD program here is like asking someone to go write a windows 10 compatible OS from scratch. You are asking for 1) something that has been done to death for decades and 2) is too much to re-create from scratch.
Thank #lastchance. This is optimizing the shape of airfoil NACA 0015 by considering max torque and we used the Differential Evolution method for a loop.
Sorry, #jonnin if that bothered you. This topic, I hoped somebody can help me know how this code work or can share some code the same. I learn C just one week, so sorry again.
Sorry, #jonnin if that bothered you. This topic, I hoped somebody can help me know how this code work or can share some code the same. I learn C just one week, so sorry again.
@TrungTrieu,
You can't "optimise the shape of airfoil NACA0015" - it has a defined shape. The 00 means it is symmetric (ie uncambered) and the 15 means its maximum thickness is 15% of chord.
In the reference that I note they change the thickness of the airfoil - which means that it ceases to be NACA0015, by definition. That journal has ceased to exist, and few people would rate its standard of reviewing.
They used ANSYS Fluent ... and I recommend you do the same.
You can't "optimise the shape of airfoil NACA0015" - it has a defined shape. The 00 means it is symmetric (ie uncambered) and the 15 means its maximum thickness is 15% of chord.
In the reference that I note they change the thickness of the airfoil - which means that it ceases to be NACA0015, by definition. That journal has ceased to exist, and few people would rate its standard of reviewing.
They used ANSYS Fluent ... and I recommend you do the same.
it didnt bother me. I am just telling you the scope of what you asked for. The problem is too large for a simple block of code suitable for the forum. I don't think I could fit a 2-d tabletop pool game (without graphics) in here, and your problem is a couple of orders of magnitude more complex.
Last edited on
@lastchance
thank you. I knew that NACA 0015 is baseline. first, you need to calculate torque after meshing in Ansys Fluent. This result link to the loop optimize, and the loop need change a thickness of airfoil and calculate torque for a new airfoil.
They used DE Algorithm for the loop, for DE code in C programming it creates random vector initial and starts to optimize.
https://www.hindawi.com/journals/isrn/2012/528418/
I don't know if we use only Ansys Fluent, we can create the loop the same? Or we will change thickness by an increase of 20%, 25% chord... Right? Create, mesh and run Ansys Fluent 5 time for 5 change thickness of airfoil?
thank you. I knew that NACA 0015 is baseline. first, you need to calculate torque after meshing in Ansys Fluent. This result link to the loop optimize, and the loop need change a thickness of airfoil and calculate torque for a new airfoil.
They used DE Algorithm for the loop, for DE code in C programming it creates random vector initial and starts to optimize.
https://www.hindawi.com/journals/isrn/2012/528418/
I don't know if we use only Ansys Fluent, we can create the loop the same? Or we will change thickness by an increase of 20%, 25% chord... Right? Create, mesh and run Ansys Fluent 5 time for 5 change thickness of airfoil?
@jonnin
I will try to focus problem necessary for next time. thank you! I hope you can help.
I will try to focus problem necessary for next time. thank you! I hope you can help.
how many things are you wanting to change here? Likely the changes affect each other, eg change chord and change wing length or whatever play off each other, and running every possible combo is probably not wise... it may not be a simple set of nested loops, you may want to do some sort of randomization and drive to the local minimums for all the variables, then re-randomize to see if there is a better minimum (or max) in another area... but yes, you are going to do some sort of problem setup, call ansys, save results, repeat, and then maybe a bit of code to compare the results to return which (one or more) was 'best'. If the number of variables is small, brute force will work, and then it would just be nested loops.
Last edited on
This is refer document: https://www.hindawi.com/journals/isrn/2012/528418/
If you have read, you can see they have 3 variables: solidity, tip speed ratio and geometry of airfoil ( airfoil cross-section) and they are fixed solidity, tip speed ratio while changing airfoil cross-section (chord is constant).
Yes, I understand this idea. Thank you so much.
If you have read, you can see they have 3 variables: solidity, tip speed ratio and geometry of airfoil ( airfoil cross-section) and they are fixed solidity, tip speed ratio while changing airfoil cross-section (chord is constant).
Yes, I understand this idea. Thank you so much.
Topic archived. No new replies allowed.