| |
An Analog Neural Network Controller for an Active Power Filter Based on the Instantaneous Reactive Power Theory
| [View]
[Download]
|
| Author(s) |
J.M. Carrasco; J.M. Quero; R. Gomez; L.G. Franquelo |
| Abstract |
A neural controller implementing a control strategy of active power filters using switching devices is proposed on the basis of the instantaneous reactive power theory. This is an alternative to implement the controller in these kind of power circuits versus complex classical analog implementations. The properties of the instantaneous power theory are analyzed and as a result, it is possible generate a surface to calculate the instantaneous active and reactive power. An analog neural network (ANN) learns to approximate the surfaces by means of a learning algorithm. A simple electronic implementation of this controller is proposed. This realization is simpler than those previously proposed based on analog multipliers and dividers. Simulation results are summarized confirming the validity of the neural network approach. |
| Download |
| Filename: | Unnamed file |
| Filesize: | 352.8 KB |
|
| Type |
Members Only |
| Date |
Last modified 2018-05-16 by System |
| |
 |