Automated Identification of Failures in Doubly-Fed Induction Generators for Wind Turbine Applications

Abstract

Remote, automated monitoring of wind generators is crucial considering that wind turbines are prone to failure due to the harsh operating environment, and difficult to access or test due to the remote location of wind turbines. The demand for condition monitoring of wind generators is increasing with the rapid growth in wind power generation. In this work, a new inverter-embedded off-line test concept for automated testing of doubly-fed induction generators (DFIG) is proposed for detection and classification of defects in the 1) slip ring-brush contact, 2) rotor winding turn insulation, and 3) stator core inter-laminar insulation. The main idea is to use the rotor side inverter for injecting test signals into the rotor winding for detecting asymmetry in the machine, whenever the DFIG is at standstill. The pattern of asymmetry in the equivalent impedance as a function of angle is analyzed for identifying the presence and type of DFIG fault at an early stage for efficient scheduling of maintenance. Experimental verification is given to show that the proposed test provides a simple means of identifying DFIG faults with high sensitivity and reliability without additional hardware.