AI-Assisted Condition Monitoring and Fault Diagnosis Methods for Power Electronic Converters

Power converters are widely applied in industry, such as wind power systems, PV systems, traction systems and electric vehicles. However, power converters are exposed to the outdoor environment are usually installed outdoors and exposed in a harsh environment, influenced by strong wind, heavy rain, high temperature, etc., and operated at high voltages and high switching frequencies. Over time, it is prone to occur abnormal and faults which will severely affect the operational efficiency of power electronic systems and the system safety. Thus, a solution to diagnose faults and predict the remaining useful lifetime (RUL) of power converters is essential, to take effective operation and maintenance measures in time and ensure the safe and stable operation of power electronic systems.
Moreover, with the development of a new generation of information technology, how to combine artificial intelligence (AI), big data and the Internet of Things (IoT) to develop an intelligent operation and maintenance (O&M) system for power converters, further improve the accuracy of diagnosis and reduce the cost of operation and maintenance, is becoming a new opportunity and challenge. This project mainly focuses on AI assisted condition monitoring and fault diagnosis methods for power converters. Using PV systems as the research object, AI-assisted methods are investigated for fault diagnosis and condition monitoring of key components including PV panels, DC-link capacitors, and IGBTs in PV systems.