Active and Reactive Power Control in Three-Phase Grid-Connected Electric Vehicles using Node-Level Capsule Graph Neural Network

Active Power (AP) together with Reactive Power (RP) control systems that operate with three-phase Grid- connected Electric Vehicles (EVs) control power exchanges between vehicles and grids and preserve voltage stability. The power quality together with voltage stability suffers from changes in the conditions of the grid and the manner of EV charging. When AP and RP power coordination is not done properly it causes the grid to experience more stress and reduces the total system reliability. To overcome these drawbacks, this manuscript proposes technique for enhancing AP and RP control in three-phase grid connected EVs. The proposed method is named as Node Level Capsule Graph Neural Network (NCGNN). The proposed method aims to enhance power distribution efficiency, improves voltage stability, and ensures effective power exchange between the grid and EVs while minimizing power imbalances. NCGNN predicts optimal AP and RP for grid-connected EVs to ensure balanced power exchange and voltage stability. Spider Wasp Optimizer-Multi-scale Hypergraph-based Feature Alignment Network (SWO-MHFAN), Adaptive Interaction Artificial Neural Network (AI-ANN), and Adaptive Neuro-Fuzzy Inference System (ANFIS) are some of the existing methods that are compared with the proposed method once it is implemented in MATLAB. By attaining an error of 1.8% for AP and RP control and preserving a high efficiency of 98.7%, the suggested approach outperforms traditional methods without reducing the stability and dependability of power exchange in grid-connected EVs.