Stator fault detection for multi-phase machines with multiple reference frames transformation

The paper focuses on a new diagnostic index for fault detection of a five-phase permanent-magnet machine. This machine has been designed for fault tolerant applications, and it is characterized by a mutual inductance equal to zero and a high self inductance, in order to limit the short-circuit current. This paper defines a diagnostic index for stator fault detection based upon the combination of information from two different reference frames. Both analytical and simulation analysis are adopted to validate the new diagnostic index. The analytical results are currently being validated by experiments on a dedicated test bed.