Hello,
I'm building a finite element model of a wind turbine blade in ANSYS. So far, I have managed to get the mass and centre of gravity to correspond to the measured values to within a couple of percent. The masses of the individual constituent materials add up to the correct values also. However, the natural frequencies and tip displacements under gravity are out by a lot more - around 15-20%. This implies that there is some issue with the stiffness properties of the materials that I am using.
In the blade design documentation the laminates are provided as sub-blocks, so a bi-axial +/-45 layer is treated as a single 0 degree layer with smeared properties. The data is also provided as planar values, so I am missing E33, nu23 etc. I have values for E11, E22, nu12, nu21 and G12.
In ABAQUS you can define an orthotropic material with a 'LAMINA' type which only requires the planar properties if you are using thin shell elements. Can a similar thing be done in ANSYS APDL?
Also, if I do wrongly estimate the out of plane properties, will this effect the behaviour of thin shell elements? If so, can anyone recommend a good method for estimating realistic out-of-plane properties?
I saw a new program for doing this work in Engineering Structures - might be worth a look - looked quite good
Hello Peter,
If you do not have any reinforcement in the 33 direction, the properties in the 33 direction will be same as those in the 22 direction as both the directions are normal to 11.
Additionally if you are trying to model each ply, it will give errors, pl try by using averaged properties of the entire thickness of the layup.
Hi,
I am Panu from Thailand. Prof Rob Dominy was my external examiner. Have you ever used NuMAD implemented by someone at SANDIA. I used this programme in my PhD research project. This programme generates a set of files support ANSYS APDL. I imported this set of files to generate a finite element model of wind turbine blade I studied.
You can ask for this programme from
http://energy.sandia.gov/energy/renewable-energy/wind-power/rotor-innovation/numerical-manufacturing-and-design-tool-nomad/
or I can send it to you directly if you want. Please send me your email.
My email is [email protected]
Regards,
Panu
http://energy.sandia.gov/energy/renewable-energy/wind-power/rotor-innovation/numerical-manufacturing-and-design-tool-numad/
Hi Peter,
First I would motivate you to try to model your structure with ANSYS Composite PrepPost (ACP). This is the new composite pre- and post-processor of ANSYS and allows you to model your structure much more efficient and more accurate than with MAPDL. By default ACP is used within the Workbench (WB) environment but it also supports MAPDL workflows (import mesh from cdb - build the laminate in ACP - export the cdb with section data).
As mentioned by Pramod, UD fabrics are transversal-isotropic materials where the 22 and 33 properties are the same. Therefore I would model the bi-axial materials as two layer (easy possible with ACP by using the material class Stackup) instead of manually evaluate the homogeneous properties of the biax.
Here are a few information about ACP and please let me know if you need further inputs.
http://www.ansys.com/Products/Structures/ANSYS-Composite-PrepPost
Best regards,
Rene
Hi Peter !
There is a very robust program called MCQ by AlphaStar Copr in Long Beach California, check their website, you can request a trial copy, it will write everything in ANSYS format and a generate the properties you are looking for...
- Badges
- Science topic