Continued Issue with 3DPolyFit

[jaysmall]

Well my issue of an unwanted point being graphed at (0 0) in altenbach's 3DPolyFit VI has not been entirely resolved. I have attached the LLB with the data I am trying to fit and plot. You will see constant filepaths in the block diagram to the data files, the names of which will have to be changed to point to the correct directory.

You will notice when that data is plotted there is an easily recognizable square of data which I would like plotted. You may also notice another at (0 0) which is approximated to almost zero. This zero point I can't seem to get rid of even though it is not in my data files. If someone could figure this out I would appreciate it. I've been staring at it for a while now. Thanks


Message Edited by jaysmall on 07-01-2008 12:34 PM

[jaysmall]

For some unknown reason, when I changed the grid that I scanned to be much finer (25x25 instead of 10x10 in the same range), I got no extra zeros. So it works right now, but if someone could still take a look at this, that would be great, because I know I'm going to ahve this problem down the line.

Plot I was looking for is attached.

[altenbach]

I don't see any zero point. I see four points that are about an order of magnitude larger than the rest. Your peak seems way too sharp for a low order 2D polynomial.

To find the max of the 2D polynomial, just get the array max of the simulated grid. If you need more precision, you could use it as a starting point for further refinement.

Note that most polynomials go to infinity for very large x and y, so I assume you want to find the max in the area of interest.

(attached is a quick modification).