Modeling and Simulation of AVR Systems in MATLAB/Simulink: Performance Comparison of PID and ANN Controllers

Abstract

This study presents the design and modeling of an Automated Voltage Regu-lator (AVR) using MATLAB/Simulink, with a comparative performance eval-uation against Proportional-Integral-Derivative (PID) and Artificial Neural Network (ANN) controllers. Voltage regulation is crucial to ensure reliable and stable power delivery under varying load conditions. In this work, a dy-namic system model was developed to represent the electrical network, fol-lowed by the implementation of control algorithms to maintain output voltage within acceptable limits. The proposed approach was assessed through exten-sive simulations under multiple operating conditions and disturbances. The results demonstrate that ANN-based control outperforms both conventional AVR and PID controllers in terms of peak time, settling time, rise time, and overshoot percentage. Furthermore, ANN provides superior stability and adaptability in handling load variations, establishing it as the most effective control strategy for voltage regulation.