Evaluating the Interturn and Eccentricity Fault Indices in the Stator Current and the Embedded Controller Signals of a Micro-DFIG

Abstract

This work presents a micro-based doubly fed induction generator (DFIG) test rig developed for fault diagnostic purposes. Experiments have been conducted by emulating the stator interturn short circuit (ITSC), rotor ITSC, and static eccentricity (SE) fault conditions on the micro-DFIG. The novelty of this work lies in the use of the various DFIG controller signals, in addition to the stator current signal which has extensively been used in previous investigations for fault detection. Another contribution is in identifying which signal is best suited for diagnosing different electromechanical faults in the micro-DFIG operating at sub-synchronous and super-synchronous speeds, while also injecting different amounts of active power into the grid. The comparative studies conducted in this work have illustrated that the stator current and the $q$-axis component of the rotor current error ($error_{Irq}$) signals were best suited for diagnosing the ITSC and SE faults, whereby the sensitivity of the fault indices in these signals was more significant at lower speeds. Overall, the $ (1-2s)f_{s}$ index in the stator current signal was identified to adequately detect stator ITSC faults at different power levels, whereas for the rotor ITSC fault, it is the $2sf_{s}$ index in the $error_{Irq}$ that is recommended for fault diagnosis. For the SE fault condition, it is the $f_{s}\pm f_{r}$ index in the stator current signal that is recommended for use at different power levels. The use of multiple fault diagnosis signals and their extended sensitivity analysis under different DFIG operating conditions significantly improves upon existing diagnosis techniques, offering more robust and reliable fault detection in DFIGs.