A Real-Time Range-Adaptive Impedance Matching Utilizing a Machine Learning Strategy Based on Neural Networks for Wireless Power Transfer Systems

In this article, the implementation of a machine learning (ML) strategy based on neural networks for the real-time range-adaptive automatic impedance matching of wireless power transfer (WPT) applications is discussed. This approach for the effective prediction of the optimal parameters of the tunable matching network and the selection of range-adaptive transmitter coils (Tx) is introduced in this article, aiming to achieve an effective automatic impedance matching over a wide range of relative distances. We propose a WPT system consisting of a tunable matching circuit and three Tx coils that have different radii and are simultaneously controlled by trained neural network models, returning an output set of matching capacitances as well as the optimal single transmitter among the three transmitters. In addition, a proof-of-concept prototype of the entire real-time range-adaptive automatic impedance-matching system is built and characterized. Finally, the proposed approach achieves a power transfer efficiency (PTE) of around 90% for ranges within 10–25 cm.