Microgrid system for electric vehicle charging stations integrated with renewable energy sources using a hybrid DOA–SBNN approach

Abstract

Read online

Microgrid-equipped electric vehicle charging stations offer economical and sustainable power sources. In addition to supporting eco-friendly mobility, the technology lowers grid dependency and improves energy reliability. The manuscript introduces a hybrid technique for efficient electric vehicle (EV) charging integrating the Dollmaker Optimization algorithm (DOA) and spatial Bayesian neural network (SBNN). This method optimizes the joint operation of photovoltaic (PV), wind turbines (WTs), supercapacitors (SCs), and battery energy storage systems (BESSs) in microgrids to enhance EV charging station efficiency, reliability, and power quality while reducing grid outages. The SBNN predicts EV load demand for improved efficiency and reliability, while DOA manages microgrid (MG) fluctuations to ensure seamless EV charging. The MG system features a four-phase inductor coupled interleaved boost converter (FP-ICIBC) and a fractional-order proportional-integral-derivative (FOPID) controller for optimal power management. An evaluation in MATLAB compares DOA–SBNN with existing approaches, demonstrating its effectiveness in enhancing EV charging performance. The proposed method outperforms all current techniques, including the Multi swarm Optimization (MSO), the Multi-Objective Gray Wolf Optimizer (MOGWO), and the Modified Multi-objective Salp Swarm Optimization algorithm (MMOSSA). The results show that the energy efficiency of the recommended approach is 19.19%, 26.15%, and 32.57% higher than the three current techniques, respectively, and that of total harmonic distortion (THD) is 19.09%, 25.85%, and 31.17% lower than those three techniques, respectively.

Keywords

• battery energy storage systems
• charging stations
• electric vehicles
• fuel cells
• four–phase inductor coupled interleaved boost converter
• photo voltaic