Integrating renewable resources and electric Vehicles: An approach for effective energy management in DC microgrid

Abstract

Direct Current microgrids (DCMGs) are vital in reducing carbon footprints and combating global warming. However, maintaining stable DC bus voltage and ensuring efficient energy flow remains challenging. This manuscript proposes a hybrid method combining Parrot Optimization (PO) and Quantum Self-Attention Neural Networks (QSANN), named PO-QSANN, to improve the Energy Management System (EMS) of DCMGs with integrated Electric Vehicles (EVs) and photovoltaic (PV) generation. This work aims to reduce operational costs and enhance system efficiency by stabilizing the DC bus voltage. PO, inspired by the intelligent behavior of parrots, optimizes the PI controller’s gain parameters for effective energy management, while QSANN uses quantum-inspired attention mechanisms to predict the load demand accurately. The proposed PO-QSANN technqiue, excluded in MATLAB/Simulink, is contrasted with existing methods such as Artificial Rabbit’s Optimized Neural Network (ARONN), Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and Whale Optimization Algorithm (WOA). Results show that PO-QSANN reduces the operational energy cost of DCMG to 0.4 $/kWh, outperforming ARONN at 0.5 $/kWh, PSO at 0.6 $/kWh, GA at 0.7 $/kWh, and WOA at 0.8 $/kWh. These findings highlight the superior capability of PO-QSANN in optimizing energy distribution and stabilizing DC bus voltage in DCMGs.