Trainning Functions
Supposing that runing a model is very costly, such as Finite Element Analysis and a simulation, an approximation Response Surface method (also called as Data Regression) is used to present the orginal model.
The following information are needed to:
- Input variables, which must be from the descision variables of an
optimization problems. The lower and upper boundaries must be set.
- A function. The function can be from
- the current optimization models.
- an executable file, such as myfun.exe (the default input file is 'input.txt' and the
default output file is 'output.txt'). The input file is used to send the value of the
input variables and the output file is used to get the results from the external
executable file.
- a dynamical link library (.dll) with a function. It is very important to make sure the
input variables in this system are matched the ones in the Dll.
- A Response Surface Model
- Number of points. The default point is 100.
To create a trainning: click the menu "Edit->Trainning",
a dialog will be shown for seting the above information.
To run the trainning procesess: click the menu
"Run->Trainning", or "Edit->Trainning".
Related Links: