Gaussian Mixture Model-Driven Scenario Analysis and Human-Centered Performance Assessment for Autonomous Photovoltaic-Battery Electric Vehicles Charging Stations

The rise of electric vehicles (EVs) and renewable integration such as photovoltaic (PV) sources has made autonomous PV-powered battery EVs (BEVs) charging stations (PVCSs) an essential solution for sustainable electrified transportation. This work presents an original study of such PVCSs, leveraging Gaussian mixture model (GMM) scenario creation, human-in-the-loop analytics, and state-of-the-art Matlab modeling. The originality consists in combining realistic behavioral scenario generation, robust state-of-charge (SOC) and load supervision, and a suite of new visual figures for transparent, user-centric academic evaluation. The “human-in-the-loop” approach, which involves incorporating real-world user behaviors and unpredictability into the computer simulation, is a significant aspect of this research. By capturing daily/behavioral dynamics, renewable energy sources fluctuation, and storage effects across five analytics figures, it has been demonstrated how smart controls and data-driven insights can enhance self-sufficiency and robustness. Results highlight the design trade-offs, detect potential vulnerabilities, and demonstrate the viability of off-grid PVCSs even under human behavioral uncertainty.