Coordinating renewable energy (RE) utilization in electric vehicle (EV) charging aligns charging schedules with the fluctuating availability of renewable energy sources to ensure efficient, sustainable power use. Nevertheless, it is extremely difficult to balance minimizing operational costs, increasing system efficiency, lowering emissions, and complying with dynamic charging requirements. To address these issues, this manuscript proposes a novel hybrid technique, Superb Fairy-wren Optimization with Frequency Pyramid Graph Convolutional Network (SFwO-FPGCN), designed to improve both economic and environmental aspects of RE-integrated EV charging. The SFwO algorithm optimizes the coordinated distribution of renewable and grid power, promoting energy balance and reduced reliance on fossil fuels. Simultaneously, the FPGCN component forecasts spatial–temporal variations in EV charging demand, enabling proactive and intelligent energy scheduling decisions. The suggested SFwO-FPGCN model was developed and tested in MATLAB and compared with various existing techniques, including Dung Beetle Optimizer-Binarized Spiking Neural Networks, Parrot Optimizer-Dynamic Weighted Hypergraph Convolutional Network, Deep Neural Network, Adaptive Neuro-Fuzzy Inference System, and Light Spectrum Optimizer-Deep Attention Dilated Residual Convolutional Neural Network. Findings indicate that SFwO-FPGCN has the lowest total harmonic distortion of 1. 39%, a low operating cost of 1546, a high efficiency rate of 98. 7%, a reduced emissions rate of 63. 9 ppm, and high accuracy in predicting demand. The method also guarantees the smooth integration of renewable and grid power sources, thereby improving the performance, reliability, and sustainability of EV charging systems powered by RE.
Venkatesh et al. (Sun,) studied this question.