Three-Level T-Type Inverter Fault Diagnosis and Tolerant Control Using Single-Phase Line Voltage
Kuei‐Hsiang Chao, Long-Yi Chang, Fu-Qiang Xu
Abstract
This study combines extension theory and chaos theory to create a three-level T-type inverter fault diagnosis system that uses single-phase line voltage signals. First, the three-level T-type inverter single-phase line voltage output waveform was measured when faults occurred separately in 12 power transistors. Subsequently, the Lorenz master-slave dynamic error transformation was used to obtain the chaos eye coordinates, which functioned as fault characteristics. Fault diagnosis involved extension theory-based fault categorization. Specifically, fault characteristic values were adopted as the input signal to determine the correlation between chaos eyes coordinates and transistor fault, thereby locating faulty transistors. In summary, this study established a low-cost and fast-operating inverter fault diagnosis system. The system integrates detection results with inverter fault-tolerant control to enable the constant operation of inverters in power generation without derating, thereby greatly enhancing system reliability. Finally, the reliability of the smart fault diagnosis system and fault-tolerant control was verified through measurement results.