OPTIMIZING ELECTRIC VEHICLE CHARGING INFRASTRUCTURE WITH EVGRIDNET BY INTERNET OF THINGS AND MACHINE LEARNING STRATEGIES.

In the arena of renewable energy integration for electric vehicle (EV) infrastructure, it is a problem to efficiently use solar power with EV charging, considering grid constraints and user preferences. The current study proposes a new smart charging algorithm that utilizes IoT data for the dynamic optimization of EV charging patterns. A novel aspect of the study was a deep learning model, "EVGridNet", that reliably predicts solar energy output and grid prices. EVGridNet uses deep learning approaches to process accumulated data via IoT devices, facilitating fine-tuned adjustments to charging patterns through predictive analytics. This algorithm receives actual data on the generation of solar energy, the price of grid electricity, and other characteristics set by the user to optimize the usage of solar energy, limit the usage of grid electricity during the peak hours and meet all the needs of the user. The optimization process of the algorithm strategically manages energy sources, uses battery storage systems to exploit solar power effectively and uses grid electricity during low-cost periods, all within user preference parameters. The proposed system has the potential of reducing grid electric use by up to 25% for EV charging, and increasing the renewable energy electricity in EV charging to 40% as per simulation results. This will enable quick EV infrastructure scalability, low carbon emission, and energy independence. EVGridNet is an outstanding innovation in smart charging technology, which is a cost-effective and scalable solution to the renewable energy sector's primary challenge. One of the key aspects of the optimization process is the control of energy sources where battery storage systems allow for flexibility in using solar energy and grid electricity within certain pre-set thresholds. [ABSTRACT FROM AUTHOR]