I have a similitude of wind blade.
The scale of length of prototype and similitude is not equal to scale of height of the cross section and the width of cross section differs from both.
How can I solve such thing because I read that everything should scaled with same scale?
I would like to suggest you that first you conduct your respective experiments with the similitude of wind blade and get some data and validate those data with cfd packages then after your next face is consider the two case first scale up the similitude geometry and make same dimensions as the prototype & simulate this case and the second case is scale down the similitude model with same factor and simulate that and analyze both the cases then you ill able to know what effect are going to play when you scale up or down geometry with same factor.
Similitude applies only to same proportion/model. You need to conduct experiment or at least gather data for a wind turbine with same proportion as your model.
interpretation with one scale factor is much more easier
litterature reports many cases with two scale factors : generally one for cross section and one for height, but only if needed : the model cannot respect one scale factor only e.g. the use of an existing installation makes impossible the one scale factor model
then in that case one should be familiar with for results interpretation
with three SF it is much more difficult, I read no papers speaking of this kind of situation
Are there dimensionless groups that apply to the situation? E.G. does the Buckingham PI theorem apply and does that permit analysis to develop guidance as to how to scale to achieve the desired result, even if linear geometric scaling cannot be followed?
The important thing is to consider what is called "Distorted Similarity" where someone has more than one "scaling factor". There is a very interesting book called "Scaling" written by Barenblatt that allows to understand more on that subject. One important article dealing with such distortion is here "Predicting the behaviour of...." but for Impact loading and large strains. Are you performing coupled FSI? My knowledge is for solid mechanics, so my advice would be general for fluid mechanics.
Another important approach, as said by Jean-Paul, is to "know" a priori what results are expected and then derive the relation based on the knowledge of the phenomena. By changing dimensions in a controlled way is possible to infer the effects. For the solid mechanics problem I would derive equation for a bent cross section and understand how changing dimensions affects the response in terms of, for example, force and displacement. It is easier when displacement is scaled by the general scaling factor and force is caled by a factor. It is also possible to simply obtain a correction factor for displacament to correct general displacements.
Hope that helps...
http://www.producao.usp.br/bitstream/handle/BDPI/32667/wos2012-5926.pdf?sequence=1
Make a new model with all the geometrical similarity....if you insist on using your old model forget about the discussion here go ahead with your test and doctor your results. another btime wasting discussion
- Badges
- Science topic
- Similar topics
- Engineering
- Thermal Engineering
- Thermofluid
- Fluid Mechanics