| INVESTIGATION AND COMPARISION OF NEURAL NETWORK APPLICATIONS TO THE CONTROL OF ELECTRICAL MACHINES | ||||||
| Author(s) | G. Henneberger; B. Otto; A. Brösse | |||||
| Abstract | In the last decade there has been a lot of interest in the research on neural networks. The main applications are pattern recognition and speech identification. Furthermore neural networks are well suited to control nonlinear systems, as shown by Narendra et Partharasathy (4). But up to now there exist only a few practical applications to the control of electrical drives. Nevertheless neural networks are not only well suited for pattern recognition but also for identification and control of dynamical systems. This paper points out following control problem: a neural network based speed controller should adapt to a servo drive, which is a nonlinear, dynamical system. This is done using an offline learning technique and the backpropagation algorithm. The results are compared with the results of another controller such as the conventional state controller. The ability of neural networks to control electrical drives is shown by simulation results. |
|||||
| Download |
|
|||||
| Type | Members Only | |||||
| Date | Last modified 2018-05-15 by System | |||||
 |
||||||