Fault Detection Method for Power MOSFET Based on Infrared Thermography

Power MOSFET has high input impedance, simple drive circuit, small drive power, fast switching speed and high working frequency. It is widely used in power electronic devices with no more than 10kW. In this paper, the theory of MOSFET power loss is introduced in brief. And a detecting method of MOSFET for boost converters based on infrared thermography is proposed in detail. Firstly, infrared images of NPC inverter are acquired based on the thermography and background of the image removed through Otsu's statistical threshold selection algorithm. Then, the RGB image is turned into a gray image. Lastly, hotspots of the image are extracted and a detecting method is proposed to diagnose whether the MOSFET is in fault or normal condition. The experimental results prove the effectiveness and feasibility of the proposed method.