A Novel Foreign Object Detection and Classification Algorithm for Capacitive Wireless Charging Systems

Wireless power transfer (WPT) systems have gained widespread adoption in recent years owing to their wide application range and high performance. However, these resonant systems are designed to operate at a fixed operating point and the presence of foreign objects may degrade their performance considerably, even potentially damaging the charger. This paper proposes a simple foreign object detection (FOD) algorithm for capacitive WPT systems using a machine learning-based classifier. This FOD technique does not require any additional sensing circuitry and uses the inherent WPT system components for sensing. A Machine Learning classifier is used to systematically categorize the foreign objects into categories. The proposed algorithm is tested with a wide set of foreign objects and is shown to have a good detection accuracy of 89%.