Diagnosis and Resilient Control for Multiple Sensor Faults in Cascaded H-Bridge Multilevel Converters

This paper proposes a unified solution for single or multiple sensor faults in the cascaded H-bridge multilevel converters (CHBMCs), including the identification of fault location and fault types, and resilient control. The original states and sensor fault information are decoupled completely and estimated simultaneously by a reduced-order observer. The accuracy and robustness of the observed variables are further enhanced by the error compensation and parameter estimation. Feature extraction from the estimated fault information identifies the location and specific types of faults. Afterward, the resilient control against sensor faults is seamlessly performed by using the original states to substitute the flawed sensor data. The experimental findings from a CHBMC prototype confirm the efficacy of the proposed approach for various scenarios of sensor faults.