Automated Fault Diagnosis Technology for Modular Multilevel Converter Based on Infrared Temperature Measurement

The continuous and uninterrupted operation between various modules of modular multi-level inverters is prone to overheating and faults. However, manual overheating fault detection has drawbacks. Thus, we analyzed the correlation between faults and temperature in modular multilevel inverters in this study. Using an infrared thermal imager to collect temperature images periodically, we analyzed temperature anomalies and change patterns in the images. We used infrared temperature measurement algorithms and correlation analysis and found that different types of overheating faults were caused by the occurrence of hot spots or temperature anomalies. The abnormal phenomena need to be monitored to determine the types of faults. The detection and isolation index were 0.971 and 0.953, and the diagnosis time was 77.94 ms. The method showed excellent diagnostic accuracy and high diagnostic efficiency. The developed technology in this study can be used to diagnose common fault types of inverters.