Adaptive energy management with machine learning in hybrid PV-wind systems for electric vehicle charging stations

This study focuses on modelling and controlling hybrid Photovoltaic (PV) and wind energy systems for Electric Vehicle (EV) battery charging stations. A load shedding mechanism based on Deep Neural Networks (DNN) has been developed. The integration of this mechanism with the grid through a voltage source converter (VSC) has been examined. The study emphasizes the integration of a DNN-based controller and maximum power point tracking (MPPT) algorithm. The proposed system aims to reduce energy costs and optimize the load on the grid by increasing the use of renewable energy sources. This innovative approach has been developed to address load balancing and stability issues in existing energy management systems. The primary goal of the research is to ensure more efficient use of renewable energy sources and to optimize energy demand management. Simulations conducted under various operating conditions have evaluated the ability of the proposed DNN controller to manage loads dynamically according to three priority levels. According to the results, the proposed system improved load shedding by 25% and increased overall system stability by 15%. The performance rate of the DNN controller was measured at 98%. This indicates that the system operates with high accuracy. In conclusion, the proposed system validates its effectiveness in managing load shedding, maintaining system stability in energy management and optimizing renewable energy sources.