Contents - Index

Curve Fit Guess Values



The Enter/Edit Equation option allows the user to enter any equation of the form Y=F(X) with up to 7 unknown coefficients represented as a0, a1, a2, ... a6.  EES will employ a nonlinear least squares curve-fitting method to determine the unknown coefficients.  The parameter estimation algorithm requires that you supply guess values for each of the unknown coefficients. Better guess values increase the likelihood of a successful curve fit and reduce computational time.  You may also supply lower and upper bounds for each coefficient if you wish to restrict the range of the search.  The guess value must be within the lower and upper bounds.


The guess values dialog window also provides fields for the maximum allowable number of iterations and the relative converge tolerance.  The search for the best set of coefficients terminates when either the maximum number of iterations is exceeded or the sum of the squares of the residuals for two successive iterations is less that the product of the tolerance and the current residual sum of squares.  After the curve fitting process is completed, you may view statistical information about the fit including the number of iterations required by pressing the Stats button.