Question: GlobalSolve: nonoptimal solution reported even when start at minimum

I am struggling with GlobalSolve (in Maple 12) and would really appreciate some help.
The problem is that a non-optimal solution is being found
Even if I start the search on an initial point at which the objective function is relatively small then GlobalSolve ultimately reports a point with a higher objective function value (despite correctly calculating the objective function at both points).
This can clearly be seen in the attached file output.pdf (pg 6), obtained by setting the info level high.
-GlobalSolve reports an initial objective function of 0.00872
-It then improves on this slightly and reports a "current optimum estimate" of 0.00871
-but then it replaces this with a HIGHER "current optimum estimate" of 0.01883
-ultimately reporting a point with an objective function value of 0.01829
I guess I'm doing something stupid but I really cant work out what...
the code (& ouput) are attached. I have also attached the data file data2.xls that is read in by the code
thanks v much
pdf file of the output:

maple worksheet that does the fitting:

View 8402_Copy of fit_all.mw on MapleNet or Download 8402_Copy of fit_all.mw
View file details

data to be fitted (read in by the worksheet)

Download 8402_data2.xls
View file details

Please Wait...