Diagnosis and isolation of air-gap eccentricities in closed-loop controlled doubly-fed induction generators

Closed-loop control provides a challenge for fault diagnosis in electric machines due to inherent compensation provided by the controller. This paper presents a method to detect static and dynamic air-gap eccentricities in doubly-fed induction generators (DFIGs) operated for closed-loop stator power control. A model of the eccentric machine is first developed using modified winding function theory (MWFT) and a suitable reference frame is chosen so as to decouple the flux equations and dynamically simulate the machine. A modified control technique is proposed to enable detection of air-gap eccentricities. The technique helps detect and differentiate or isolate this fault from grid-voltage unbalances and stator and rotor electrical faults that produce similar effects. The proposed method looks at low frequency fault signatures which can be detected using the DSP controller that is part of the drive. The effectiveness of the proposed control is verified using simulations and preliminary experiments performed on a laboratory set-up.