Optimized Modeling and Evolutionary Control of Three-Phase Magnetic Resonance Wireless Power Transfer Chargers with Artificial Neural Network–Based State of Charge Estimation for Heavy-Duty Electric Vehicles

Wireless chargers for electric vehicles have recently attracted considerable attention from researchers and scientists, particularly due to their promising applications for high-power transfer, such as in heavy-duty vehicles, which are the backbone of transportation. This paper presents an innovative method for controlling a threephase magnetic resonance wireless power transfer charger using genetic algorithm optimization of parameters to ensure efficient wireless energy transfer. This control method, which operates exclusively on the primary side, regulates the battery current in the case of charging using the constant current-constant voltage method and also in the case of charging using the multi-step current method, which preserves the state of health of the battery. The state of charge (SOC) of the battery is estimated by neural networks, which ensure SOC estimation despite communication loss. The proposed control method is tested under several scenarios for a wireless power transmission of 20 kW with a resonance frequency of 85 kHz and an efficiency of 90.15%. The results obtained demonstrate the robustness and effectiveness of the proposed method for transferring wireless power to heavyduty vehicles despite misalignment between the coils and variation in the air gap, two frequent and disruptive phenomena in real-world situations.