Lyapunov-based training algorithm applied to a continually on line-trained ANN used in the current-loop control of a single-phase switched rectifier

Abstract:

This paper presents an implementation of a PWM single-phase switched rectifier controlled by a continually online-trained artificial neural network (COT-ANN). The stability of the COT-ANN training is ensured by using a suitable description of the switched rectifier and a Lyapunov-based training algorithm. The stability of the neural network is verified using a norm metric of the ANN matrix weights. The proposed switched rectifier can reverse the power flow direction while attaining power factor regulation. Simulations are used to test the validity of the proposed algorithm and the results are finally verified by a practical implementation of this system. Copyright © 2007 John Wiley & Sons, Ltd.

Año de publicación:

2008

Keywords:

• power electronics
• Lyapunov methods
• Neural networks applications
• COT-ANN training
• rectifiers

Fuente:

scopus