Smart Adaptive Vehicle-to-Grid and Grid-to-Vehicle Control for EV Charging Stations Using Artificial Intelligence and Neuro-Fuzzy Systems.
POTNURU ESWARA RAO 1, PILLA JAYA 2, BATTA LOKESH 3, KEERTHI PAVANI 4,
KUPIREDDI BHUVAN 5, UPPULURI LAKSHMI 6
1EEE STUDENT & SANKETIKA INSTITUTE OF TECHNOLOGY AND MANAGEMENT
2EEE STUDENT & SANKETIKA INSTITUTE OF TECHNOLOGY AND MANAGEMENT
3EEE STUDENT & SANKETIKA INSTITUTE OF TECHNOLOGY AND MANAGEMENT
4EEE STUDENT & SANKETIKA INSTITUTE OF TECHNOLOGY AND MANAGEMENT
5EEE STUDENT & SANKETIKA INSTITUTE OF TECHNOLOGY AND MANAGEMENT
6 Assistant Professor & SANKETIKA INSTITUTE OF TECHNOLOGY AND MANAGEMENT
ABSTRACT
The rapid growth of electric vehicles (EVs) has increased the demand for efficient and intelligent electric vehicle charging stations (EVCS). This research presents the development of an Artificial Intelligence (AI)-based adaptive Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) controller for a DC microgrid-based EV charging station integrated with a solar photovoltaic system (SPVS), storage battery (SB), electric vehicle (EV), and utility grid. The proposed system is designed for residential buildings and office environments where EVs remain parked for extended periods and can be utilized for intelligent energy management. An Adaptive Neuro-Fuzzy Inference System (ANFIS) combined with Artificial Neural Networks (ANN) is proposed as the power management controller (PMC) to enhance system performance and decision-making capability. The controller dynamically manages the power flow between the SPVS, storage battery, EV, and grid depending on power availability and load demand. The system operates in two distinct modes: Vehicle-to-Grid (V2G) mode, where the EV supplies power to the building or microgrid during power shortages, and Grid-to-Vehicle (G2V) mode, where the EV is charged using power from renewable sources, battery storage, or the grid. When the power generated from the SPVS and storage battery is insufficient to meet the load demand, the controller intelligently extracts power from the EV through V2G operation. If the available renewable and EV power are still inadequate, the deficit power is supplied by the grid using G2V operation. The proposed Neuro-Fuzzy based power management controller maintains a stable DC bus voltage and improves system dynamic performance The results confirm that the proposed intelligent control strategy enhances power management, improves system stability, and increases the efficiency of EV charging stations integrated with renewable energy sources.
