Tool for the design and implementation of control systems with neural networks

This document shows a very versatile application for the design and implementation of neural controllers in simulation. This application runs over Matlab Software. The application allows to follow all the steps in a process of designing control systems. Processes of identification can be made. At the beginning a model of the system that wants to be controlled is built. Entry data adapted to a system is generated and results at the exit are obtained. With the database the deduction of the model of the plant is made and design of neural control inverse model type: general o specialized. Then the controls can be tested directly in the application or exported the neural controller to Simulink. This characteristic of the application allows to test the controls with real systems through an interface that combines applications of Matlab and Labview.