Portal de Periódicos da CAPES

ESTRATEGIA DE RE-ENTRENAMIENTO DE REDES NEURONALES PARA MEJORAR EL CONTROL DE LA EXCITACIÓN DE UNA MÁQUINA SÍNCRONA

2006; Volume: 3; Issue: 32 Linguagem: Inglês

10.22517/23447214.6199

2344-7214

Sandra Milena Pérez Londoño, Juan José Mora Flórez, Germán Morales-España,

Advanced Control Systems Design

Artificial neural networks (ANN) offer an alternative to implementation of conventional controllers due the good performance to identify and control nonlinear systems at high computational speed. This technique was used in the supplementary controller implementation (power stabilizer) aimed to maintain an acceptable behavior of the terminal voltage in a synchronous machine, when it is under disturbance conditions. In this paper the obtained results of the proposed approach are presented. The advantages of the use of neural nets over the classical adaptive implementation of power stabilizers are shown.