Modeling the Automatic Voltage Regulator (AVR) Using Artificial Neural Network

Abstract

The artificial neural network (ANN) has been successfully applied to many systems’ models with the objective of improving their performance. This study presents the ANN application to the automated voltage regulator (AVR) model of the synchronous machine. The studied system consists of the single-machine infinite bus (SMIB) power system. Data was extracted over a diverse range of conditions. Different fault types were also considered for collecting a large number of data to avoid overfitting. The multilayer feedforward neural network (MLFFNN) approach was applied to design the paradigm and system equations were solved using the MATLAB Simulink software.