MACHINE LEARNING DRIVEN CLOUD EDGE ADAPTIVE CONTROL FOR HIGH-FREQUENCY WIDE-BANDGAP EV CONVERTERS

ABSTRACT In the article, the authors presented a machine learning-inspired cloud-edge adaptive control system of the high-frequency wide-bandgap (WBG) EV converters. The suggested solution integrates power electronics, which is outlined on the principles of the implementation of the SiC/GaN with the real-time edge inference and the optimisation in the cloud as the means of ensuring that the efficiency, thermal stability, and dynamic response are optimised and that the challenges of scalability and latency of intelligent EV power systems are addressed. Keywords: Wide-bandgap converters, cloud-edge computing, machine learning control, electric vehicles, adaptive power electronics.