Electrical signature analysis based online monitoring of drive-trains for doubly-fed wind generators

Drive train failures are one of the common failure modes of wind turbines. Their early detection, including the generator bearing and gearbox defects, is considered difficult using the state-of-the-art monitoring techniques. In this paper, a novel electrical signature analysis-based drivetrain monitoring technique is proposed for wind turbines. A novel electrical signature tool, electrical multi-phase imbalance separation technique (eMIST), is proposed to improve the signal-to-noise ratio in electrical signature analysis. The theoretical basis of drivetrain defect detection is also presented in detail. The proposed approach is validated by experimental results obtained from a 25 HP wind drivetrain simulator, designed to simulate 1.5 MW wind turbines. The experimental results show that the proposed approach is capable of providing accurate detection of drivetrain defects at an early stage. The proposed approach is cost effective with high probability of detection (PoD) of drivetrain defects compared to existing techniques.