Detection of Inter-Turn Short-Circuit on a Doubly Fed Induction Machine With D-Q Axis Representation - Application to Different Power Levels

Abstract

This paper presents a method based on the elliptical representation of D-Q currents to detect and quantify an Inter-Turn Short-Circuit (ITSC) fault in windings of a Doubly Fed Induction Machine (DFIM). ITSC is said to be an evolving fault, so it is essential to detect it at an early stage to avoid damage on the machine. Therefore, the method should be able, on the first hand, to detect the defect and, on the second hand, to quantify its severity. Moreover, this study requires less computation time than classical methods such as harmonic analysis. In this paper, current data are acquired at a sampling frequency of 1 kHz. This method is successful with this low data sampling rate. In order to validate this study, a theoretical analysis with two models of different DFIM powers (0.3 kW, 0.25 kW and 11 kW) is carried out (healthy case and faulty case: presence of ITSC), and these results are confirmed by using platforms including Doubly Fed Induction Machines (DFIMs) and Data Acquisition (DAQ) system.