Solar-Powered EV Charging Using Modified SEPIC-Luo Converter with Recurrent Neural Network Technique

In order to address issues with renewable energy utilization processes and the growing power consumption of Electric Vehicles (EVs) in the near future,a solar powered charging station for EV is developed. Initially, a highly efficient Modified Single Ended Primary Inductor Converter (SEPIC) Luo converter is used to increase a low voltage of the PV system. The maximum power of the Photovoltaic (PV) system is then tracked using the Recurrent Neural Network based Maximum Power Point Tracking (RNN-MPPT), whose parameters are adjusted using the Monarch Butterfly Optimization (MBO) algorithm. Then, the high frequency full bridge inverter effectively transforms the power and isolation transformer is utilized for decreasing electrical noise and interference. Furthermore, the interleaved synchronous rectifier is used for attaining effective charging by reducing conduction losses. The developed work is applied in Matlab/Simulink software, reveals that the developed work attains the converter efficiency of 97.44%when compared to 90% of Luo and 95.16% of Enhanced SEPIC, ensuring the stable and reliable power delivery. Also, The MBO-RNN approach exhibits 98.8% of tracking efficiency and a root mean square error of 0.0125.