Does any one know how to evaluate the porosity in a given structure as shown in the attachment? Suppose we do not know the parameters of all the squares.
Thank you for your reply. I may do the experimental work later. But right now, I am doing the numerical simualtions in COMSOL, and I want to evlaute the porosity values in COMSOL. Any idea about that?
Well, it is up to you to first define porosity. I deal with acoustics, and for me porosity would entail a ratio between the open area (the holes) to the total area; both of which you can easily calculate in COMSOL Multiphysics.
The definition of the porosity in your case suits mines well. I am not so good at COMSOL. Would you please give me a more detailed instruction, like how to calculate the open area? Thank you in advance.
In COMSOL Multiphysics you have several options when it comes to evaluating geometrical properties such as lengths, areas and volumes. In your case I would go to Component1->Definitions and right-click to select Component Couplings->Integration. This gives you an operator which can be used for integrating pretty much what the heart desires. Select the domain with holes in it and note the Operator name, which is probably "intop1". Go to Definitions->Variables (or right-click on Definitions if you do not see Variables in the tree, and then select Variables to have it). Create a new variable called Area and define in as intop1(1), since in your case you simply want to integrate the number 1, to get the area. If the simulation has already been run, simply Update Solution under Study, otherwise run. Now you can evalute the variable Area by including a Global Evaluation under Derived Values (we are down in Results now in the tree). And the area of the total surface minus the holes shows up in a table. For your geometry an analytic expression gives you the total area (side length x side lenght), and you can now find the ratio between the two.
A simpler method is to go to geometry in both the tree and the top bar, select the Domain view in the Graphics window, and click the ruler to get the area. For your case it will probably give the same result, but more complex geometries, this is not as accurate as the above method.
You may also consider the fact that for periodically spaced holes a representative sub-area or cell encompassing each hole can perhaps be found, so you simply consider that cell instead of the entire structure, since some times holes placed near the outer rim distort the ratio, and you may be interested in the more general ratio. This is illustrated in Fig. 2 in (http://ws.binghamton.edu/me/msl/PDF_Paper/2005%20Jan%20Viscous%20Damping%20sensors%20and%20actuators_Paper_pdf%201.pdf), where a hexagonal pattern is shown. The area of the hexagonal and the area of the circular hole can both be found analytically, so that you do not need to evaluate areas numerically.
Thank you for your reply. But I have met a problem while doing the integration of the holes' area. Since the geometry was created by other software and then imported into COMSOL, there is only one domain actually. I have no idea how to select the domain with holes.
As mentioned above, I suggest finding the area without the holes using COMSOL, and then analytically find the total area, since your figure shows a very simple geometry. If it is more complex, you may have to close off the holes or use some other strategy. If so, you can perhaps share the file (COMSOL 5.1 or 5.2 format)?
Here you go (see mail). The two areas are found numerically (with some geometry tricks to get the holes) and a ratio is given in the Results section :-)