Metal Foreign Object Detection Algorithm Based On Multivariate Normal Distribution For Wireless Power Transfer System

Aiming at the need of metal foreign object detection in wireless charging system, a metal foreign object detection algorithm based on multivariate normal distribution was proposed in this paper, which solves the problems of low detection sensitivity and blind area of single monitor coil in traditional method. This method has the advantages of low cost, high sensitivity, and strong applicability without changing the hardware circuit. First, the mutual inductance coupling model between metal foreign object and monitor coil was established. Also, the detection principle based on the variation of monitor coil impedance and multiple normal distribution algorithm were interpreted. Then, the corresponding monitor coil, circuit and signal processing were designed. Finally, the experimental verification was carried out on a 3.5 kW wireless charging prototype. As for 1-jiao coin at the position of the corner of the monitor coil, the eigenvalue output by the algorithm changed by 75.51 times compared with that without foreign objects. Thus, this algorithm can significantly improve the detection effect of small size foreign objects and which are in corner area of monitor coil.