modelling a random distribution of pores on a grid

[padbadger]

i have a little problem that i want to write a computer program to solve, but unfortunately I can't program and I was hoping someone could help. I want to randomly distribute circular objects which are 30nm in diameter (they are pores in real life) over a grid (modeling a filtering membrane), with a denstiy of 6 X 10^8 pores/cm^2, then figure out some statistics from the resulting array - such as average inter-pore distance - plus a histogram type plot of inter-pore distance against frequency of occurance.
I will be very grateful for any help.
adam

 

[rhmerrill]

How large is the filtering membrane in cm? Is it square, round, rectangular, eliptical...? If I am reading this correctly -- I think one realitively simple way of doing this is to randomly generate a series of points NxN=N2 (the two space of natural numbers). And then compare their distances using the standard point distance formula D=|x-y| (in this case x and y are vectors in 2-space). If I am understanding you correctly however (entirely possible that I'm not), the average inter-pore distance for any given pore its average distance from all other pores...One problem I see here is that this method could get really hairy as the number of pores increases. You would have to take the distance of each pore from every other pore, so the more pores you have -- the larger the number of calculations. The other problem I see is that most random number generators assume a probability distribution or use a somewhat predictable algorithm to begin with, so your ramdomly generated points, aren't truely random. Anyways, its certainly possible, but it would be nice to know more details about the application -- i.e. what question are you really looking to answer. Hope this helps in some way.