Higher dimensional waves, how to go around?

Hi,

This is probably a more algorithm question than igor and I apologize if discussed before as I couldn't find any relevant posts.

To be specific, my model has 9 independent free parameters, a1, a2, ....a9 and I wish to find the best match between model predictions and measurement at 100 discrete positions. As there are many local minima, fitting results are highly dependent on the initial guess parameter values so I want to do a grid search. I would vary a1 from 0.1 to 1.0 at a step of 0.1, a2 from 0.05 to 0.5 at a step of 0.05, and so forth. Then the question is how to store this 9-dimensional matrix as model predictions and later find the best match with measurement? If my model only had 4 free parameters, I could make a 4-dimensional wave, but how about 9-, any way to go around?

Thanks!

zh