3d data interpolation

[WERT]

I have a set of 3d raw data. Is there a VI that will simply
interpolate a z value based on the raw data if both the x and y
are inputted. Currently I am fitting a 3rd order polynomial to the
data and using this to solve my problem. The main problem with this is
that I don't know how well the polynomial fits my data. If it
outputted an r^2 value than i would be happy.

Re: 3d data interpolation

[altenbach]

Do the x,y values form an equally spaced regular grid or are the
points randomly placed.

Do you need to interpolate in one direction, e.g. along each plane
with constant x, or do you need to form a solid surface, potentially
calculating Z for any fractional x and y.

Do you know anything about the data? Is there a theoretical model that
you can fit to it? This would be the most reasonable approach in
general.

If you simply need to interpolate without any theoretical model, you
could do a 2D spline. Since a spline goes through all the data points,
chisquare will be zero.

Or are you looking for a solution that does some smoothing of your
data?

Re: 3d data interpolation

[WERT]

The x,y values form a relativly linear surface that twists in the z
direction.

The data is taken from a valve calibration. The pressure is measured
for each position of the valve. However there is a 3rd variable in the
system. This means that the pressure required is dependent on valve
position and the 3rd variable. This third variable causes small
changes in the other two.
What I am trying to do is calculate my valve position for a known
pressure and 3rd variable.

The data doesn't need to be smoothed.