Performance enhancement of EV charging stations and distribution system: a GJO–APCNN technique

This paper proposes an efficient energy management between the distribution system and electric vehicle charging system with a hybrid technique. The novelty lies in the joint execution of Golden Jackal Optimization (GJO) and Attention Pyramid Convolutional Neural Network (APCNN). This leverages the strengths of both techniques. GJO provides a good initial solution, while APCNN refines it with its data-driven learning capabilities. This technique analyzes system data and grid conditions to find an optimal control signal for the inverter which improves performance and results in reduced system cost. The primary goal of the proposed method is to improve the performance of the charging station and the distribution grid while reducing the overall system cost. The GJO method is utilized to optimize the control signal of the inverter. Develop a model that translates the control signal for the inverter into a format compatible with the GJO algorithm. The charging station is then provided with intelligent control and decision-making abilities through the application of the APCNN approach. After that, the proposed framework is implemented into practice using the MATLAB platform, and its results are compared to those of the current approaches. The proposed method shows better results compared to existing methods like Latent Semantic Analysis, Fertile Field Algorithm, and Salp Swarm Algorithm. From the result, it is concluded that the proposed technique reduces system cost more than the existing techniques. The 95% attained efficiency indicates a significant performance improvement with the proposed strategy.