Intelligent Voltage Sag Compensation Using an Artificial Neural Network (ANN)-Based Dynamic Voltage Restorer in MATLAB Simulink

Abstract

An innovative Dynamic Voltage Restorer (DVR) system based on Artificial Neural Network (ANN) technology, implemented in MATLAB Simulink, accurately detects, and dynamically restores voltage sags, significantly improving power quality and ensuring a reliable supply to critical loads, contributing to the advancement of power quality enhancement techniques. Voltage sags are a prevalent power quality concern that can have a significant impact on sensitive electrical equipment. An innovative approach to address voltage sags through the operation of a Dynamic Voltage Restorer (DVR) based on Artificial Neural Network (ANN) technology. The proposed system, developed using MATLAB Simulink, leverages the ANN's capabilities to accurately detect voltage sags and dynamically restore the voltage to the affected load. The ANN is trained using a comprehensive dataset comprising voltage sag events, enabling it to learn the intricate relationships between sag characteristics and optimal compensation techniques. By integrating the trained ANN into the DVR control scheme, real-time compensation for voltage sags is achieved. The effectiveness of the proposed system is rigorously evaluated through extensive simulations and performance analysis. The results demonstrate the superior performance of the ANN-based DVR in terms of voltage sag detection accuracy and restoration precision. Consequently, the proposed system presents an intelligent and adaptive solution for voltage sag compensation, ensuring a reliable and high-quality power supply to critical loads. This research contributes to the advancement of power quality enhancement techniques, facilitating the implementation of intelligent power system.