Signal processing-based switch fault detection methods for multi-phase interleaved boost converter
Résumé
Multiphase Interleaved Boost Converter (MPIBC), for use in high performance vehicle application, is of a great interest. It allows the transfer of energetic flows between main and uxiliary sources (fuel cells, batteries, supercapacitors) and charges (dc-bus, motors, wheels and auxiliaries) with reduction in current source ripples. In order to enhance its efficiency, Silicon Carbide (SiC) semiconductors are used instead of Silicon (Si) ones. MOSFETs and diodes used in the MPIBC are exposed to high electrical and thermal stresses. A failure of these switches strongly affects performances of the system. Therefore, switch failures handling is very important to overcome several risks in early stage and to ensure continuity of service. It consists on fault detection (first step) and fault tolerant actions (second step) to restore healthy state in case of switch short-circuit and open-circuit faults. The suggested switch fault detection methods are simple and nonintrusive based on the online monitoring of switch drain-to-source-voltage VDS and comparing it to a threshold voltage which is calculated according to a simple strategy. Simulation results show the effectiveness of the proposed method.