Stator fault detection of doubly-fed induction generators using rotor instantaneous power spectrum

This paper introduces an advanced monitoring and diagnosis technique for the detection of incipient stator electrical faults in doubly-fed induction generators (DFIG) used in wind turbines. Thanks to the measurability of DFIG rotor side electrical characteristics, a new diagnostic method based on the frequency analysis of rotor instantaneous power spectrum is proposed. Theoretical analysis shows that, compared with the common current or voltage spectrum analysis methods, the instantaneous power spectrum also has the advantage that external sensors and hardware devices are not needed. Moreover, its characteristic frequency is 2f without slip s (f is the fundamental frequency), so this method has its unique advantage which has good interference immunity to avoid the mis-diagnosis. Once some slip tracking or calculation errors occurred, the characteristic frequency would not change.