Rotor Fault Detection of Wind Turbine Squirrel Cage Induction Generator Using Teager–kaiser Energy Operator

Abstract

This paper proposes the use of the Teager-Kaiser energy operator (TKEO) for the detection of a broken rotor bar (BRB) in a squirrel cage induction generator, which is used usually in wind energy systems. The TKEO is investigated using the non-stationary signals representation in the time/frequency domain by sensing the estimation of the instantaneous amplitudes and frequencies present on the stator of wind turbine. This situation of the rotor bar breaks is caused mainly by combinations of mechanical and electrical stresses, including vibration, pendulum, thermal and centrifugal and rarely it can appear due to the fabrication and the quality of the used material's origins. Consequently, the case of a BRB may occur under these circumstances and it is a common fault which can occur in the rotor of the squirrel cage induction generator, it can be estimated to be around 20% among the eventual faults of this kind of machine. The necessary properties for the development of this representation based on the TKEO in time and frequency are presented, which can improve the diagnosis of the studied wind energy system. The obtained results of the presented method have been experimentally validated by robustness tests, which raise the effectiveness of the proposed approach, including a 4-kW induction generator for online rotor bar fault detection using the spectrum analyser and stator current sensors board designed especially for this experimental purpose.