EXTERNAL OBJECT DETECTION IN WIRELESS POWER TRANSFER FOR EV

Abstract

Wireless power transfer (WPT) is short-distance magnetic coupling between transmitter and receiver coils that allows electrical energy to be transferred without a direct wired connection, which consequently offers a safer environment. However, there are serious risks associated with external metallic objects (EMOs) in the system's operating area, such as the possibility of causing fire from overheating brought on by eddy currents. By putting forth a unique sensing coil-based External Object Detection (EOD) technique, this thesis tackles the crucial problem of EMO detection. The method uses five open-circuited single-turn sensor coils to track changes in electromagnetic parameters brought on by EMO-induced eddy currents, including resistance, mutual inductance, and self-inductance. The technique ensures safety prior to high-power charging in WPT system specifically for EVs by assessing these variations and accurately identifying EMOs while running at pre-startup power levels. In order to assess the system's reaction to various EMO scenarios, such as different shapes (such as coins, cans, and spoons), sizes, and frequency spectrum, the research combines mathematical modeling with Finite Element Analysis (FEA) simulations using Ansys Electronics Desktop. The method's robustness is confirmed by experimental validation, which was carried out over a frequency range of 80-90 kHz and shows strong detection accuracy under misalignment settings. Reliable detection of EMOs is made possible by the results, which reveal sequent error values below 1 % in EMO-free situations and escalating to 3-12 % in EMO-present cases. The study also investigates EMO localization via spatial error mapping, demonstrating increased sensitivity at higher elevations and central locations. The transmitter-side exclusivity and power-level invariance of the suggested approach make it easier to deploy in stationary EV charging infrastructure. This work enhances the practical deployment of WPT systems by improving safety and reliability, opening the door for wider implementation in sustainable transportation. To further enhance dynamic performance, future developments might incorporate hybrid sensing approaches or machine learning for adaptive calibration.