Fault detection and fault severity calculation for rotor windings based on spectral, wavelet and ratio computation analyses of rotor current signals for a doubly fed induction generator in wind turbines

Abstract

Doubly fed induction generators (DFIGs) are negatively affected by inter-turn short-circuit faults (ITSCFs) of rotor windings, and considering the safety and continuous operation of the wind turbine surveillance of DFIGs is of immense importance. For this purpose, an experimental platform was designed to detect ITSCFs in rotor windings. An experimental set-up with a 100 kW three-phase wound rotor induction machine and a condition monitoring system was constructed to obtain real-time experimental data. Different tapping points were set up to induce different ITSCFs. The rotor current was used for the analysis of the ITSCFs. First, spectral analysis was performed, and the ensuing results suggest that some harmonic frequency components were generated during the ITSCFs in addition to the fundamental frequency component. Additionally, the RMS value of the rotor current was measured during the process. In the next step, a wavelet coefficient of detail level 9 was found to be effective in analysing such harmonic frequency components in the rotor current. In the end, a new method based on the RMS value of the wavelet coefficient and the RMS value of the rotor current was proposed that can calculate the fault severity. It is concluded that such signal processing techniques are appropriate to detect ITSCFs and determine the fault severity in rotor windings in DFIGs.