Parameterized Models of Double-D Coils for DWPT Applications Through Deep Learning Techniques

Double-D (DD)-coils are essential components in dynamic wireless power transfer (DWPT) systems as they are bipolar in character, allowing a smooth flux passage through them. This paper presents a deep-learning model that helps predict the coils' mutual inductance, the coupling coefficient with and without lateral and angular offsets, and other design parameters. It is shown that the proposed deep-learning models work orders of magnitudes of times faster than conventional numerical software.