Development Of Foreign Object Detection Device For Vehicle Wireless Charging

With the continuous depletion of petroleum energy and the aggravation of environmental pollution,the electric vehicle industry has received strong support from various governments and is in a period of rapid development.The wireless charging transmission coil,as a carrier for the charging energy supply of the electric vehicle,is a key part of the wireless charging technology.Because its working environment is generally outdoor,various foreign objects such as coins,cans,and fallen leaves may appear on the surface at any time.This will seriously reduce the working efficiency of the transmitting coil during work and bring about safety accidents.Therefore,an automatic foreign matter detection device is urgently needed to realize the recognition of the foreign matter on t he surface of the transmitting coil,to ensure the charging efficiency and charging safety in the charging process,and to promote the development of the electric vehicle industry.In order to realize the automatic detection and identification of the forei gn matter on the surface of the transmitting coil,the dual sensor of the camera and the coil probe is used to monitor the surface of the transmitting coil and the distribution of the magnetic field at the same time according to the characteristics of the foreign matter detection task,and the image characteristics and foreign matter information of the transmitting coil are analyzed.The extraction method and differential coil differential principle have laid a theoretical foundation for the detection of fo reign materials with dual sensors.In order to achieve the extraction of foreign objects,through image filter grayscale preprocessing,the color tracking method of the area to be detected,and edge detection based on Canny,using the background as the difference technology to separate the foreground and background,and through the image moment principle,the foreign matter was obtained.The geometric information provides information for the following classification of foreign objects.In order to identify incoming foreign objects,a machine learning model was established.By tagging collected foreign object pictures,they were sent to the SVM model under the Open CV framework for learning.The SVM-based foreign object recognition network was trained to realize the foreign matter,spot and Light and shadow recognition improves the accuracy of the algorithm.Machine vision-based foreign object detection.In order to realize the accurate recognition of metal foreign objects,the principle of electromagnetic induction and electromagnetic coupling was analyzed.Through balance coil technology and digital filtering technology,accurate and stable differential voltage values were obtained.The use of a loop to learn voltage fluctuations without foreign body status to ensure the reasonable choice of alarm threshold.At the same time,adaptive adjustment ensures that the differential voltage is at the optimum value under different operating conditions.In this way,metal foreign object recognition based on electromagnet ic induction is realized.In order to detect the actual effect,a foreign object detection device was developed.The device is mounted on a transmitting coil and can effectively recognize foreign objects in different environments and has high recognition accuracy,which can guarantee the efficiency and safety in the wireless charging process.