Implementation of Neural Network-Based Control System for MT-Upqcμg to Enhance Power Quality

Abstract

The integration of distributed generation and renewable energy sources in Microgrids introduces significant challenges in maintaining Electrical Power Quality (EPQ) and ensuring system stability. This paper addresses the mitigation of EPQ issues, such as voltage sags, swells, and harmonics, by incorporating a Multi-Terminal Unified Power Quality Conditioner (MT-UPQC) into a hybrid photovoltaic–wind–battery Microgrid system. The effectiveness of the PV-wind-battery-UPQC system is demonstrated through MATLAB/Simulink simulations. In this setup, the shunt controllers leverage the PQ theory, whereas the series controller utilizes the Unit Vector Template approach. Results reveal notable enhancements in power quality, characterized by substantial reductions in harmonic distortions and voltage fluctuations compared to conventional techniques. Moreover, the implementation of an Artificial Neural Network (ANN)-based control strategy enhances system stability across diverse operating conditions, thereby providing a dependable solution for maintaining superior power quality within complex microgrid environments. Nonlinear, unbalanced loads and harmonic supply voltages are used to assess the performance of the Renewable energy systems combining solar PV, wind power, and battery storage with grid integration, which is controlled by an artificial neural network. MATLAB/Simulink simulations demonstrate the efficacy of integrating PV, wind, and battery systems with a Unified Power Quality Conditioner (UPQC). The control strategy involves PQ theory for shunt controllers and Unit Vector Template Generation for series controllers. This approach yields substantial power quality enhancements, minimizing harmonic distortions and voltage instability issues. Moreover, Artificial Neural Network (ANN) control enhances system resilience and adaptability, ensuring reliable operation across diverse microgrid scenarios.