I spent some time modifying the non-linear Lev-Mar fit to do a rational polynomial fit of z=f(x,y). The one I ended up using was the ratio of two polynomials, where both polynomials went up to a power of 4 in x and/or y. It worked really well for smooth functions. It is just like a least squares fit, but it uses non-linear equations and is iterative.
At the moment, I think I spent too much time on it to give it away. This might help you get started in the right direction, though. Bruce ------------------------------------------ Bruce Ammons Ammons Engineering www.ammonsengineering.com (810) 687-4288 Phone (810) 687-6202 Fax -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Max Harbor Sent: Monday, March 08, 2004 5:44 PM To: [EMAIL PROTECTED] Subject: [A] Evaluating a 3D Prametric Surface I have assembled data taken in a series of stepped tests that I have plotted using the 3D Parametric Surface Graph. Essentially the inputs are X, Y, and Z matrices that describe a ribbon like surface that does not fold under or back upon itself. I want to perform a fit and evaluation on the data such that I can input scalar X and Y and output scalar Z. Because the data are distributed unevenly I am having fits coming up with an evaluation routine to handle this. Does anyone have any suggestions or a canned solution to this kind of a problem. Thanks, Max Harbor [EMAIL PROTECTED] __________________________________ Do you Yahoo!? Yahoo! Search - Find what you're looking for faster http://search.yahoo.com
