TransGA Guide

Main Features

TransGA is a Windows application, and the standard Windows methods for executing the application, opening documents an managing windows can be applied.

The steps required for obtaining a dynamic model with TransGA are:

	Obtain a dynamic data set
	Introduce the data set in TransGA
	Modify the algorithm’s parameters
	Execute the algorithm
	Get results

Obtaining a Dynamic Data Set

TransGA obtains the dynamic model of the process by fitting dynamic data of the process measured experimentally. The experiment consists in stimulating the inputs of the process, and measuring the outputs.

Introducing the Data Set in TransGA

The input data is introduced in the table of the main window of the document. There are several ways of doing this:

	Data can be written directly in this table. First of all, the table is resized by introducing the appropriate number of inputs and data points in the edit boxes placed above the table, and then data can be written in the cells of the table. It is possible to add o delete intermediate rows of the table by using the “Add” and “Del” buttons which are above the table.
	A spreadsheet such as Excel can be used to write the data table, and then this table can be copied and pasted with the commands of the edit menu.
	Data can be loaded from a file with the commands of the file menu. This file could have been saved in a previous session with TransGA, or could have been written directly by the user.

Modifying the Algorithm's Parameters

The parameters of the algorithm can be adapted to a particular problem. These parameters can be edited with the “Options” dialog, which is opened with the command “Options” of the algorithm menu, or the corresponding button of the toolbar.

The parameters of the algorithm are:

	Population size
	Number of generations
	Population refresh period and percent
	Type and probability of reproduction
	Probability of Mutation
	Elitism
	Size reduction factor

Executing the Algorithm

After introducing the experimental data and setting the parameters of the algorithm, the algorithm can be executed with the “Execute” command of the Algorithm menu, or with the corresponding button of the toolbar.
During the execution of the algorithm, the results obtained so far will be shown in the “plot” window.

Getting Results

The final results obtained by the algorithm will be shown in the “plot” window at the end of the execution of the algorithm.

These results are also stored in a text file named NameFile_Report.txt, which will be stored in the folder of the data file. The report contains the following data:

	A summary of the parameters of the algorithm.
	The variation of several parameters during evolution. These parameters include: mean and standard deviation of the fitness of the population, number of mutations and reproductions, and best solution’s genotype, phenotype and objective function.
	A summary of the results obtained by the algorithm.

If the “show report” option is checked in the Options Dialog, this report will be automatically opened at the end of the execution.

More details

For more details, please refer to the help file of TransGA. This help file can be opened with the command "Help Contents" of the Help menu.