基于希尔伯特-黄变换的双馈异步风力发电机定子故障诊断研究

摘要

当双馈风力収电机定子绕组収生轻微匝间短路时，故障引起的信号变化十分微弱，三相定子电流的时域波形变化微小，而根据希尔伯特-黄变换(HHT)分析获取信号的时频表示，在故障时刻的突变很明显，从而提取故障特征。在动模实验室进行实验，获取定子在匝间短路瞬变过程的电流信号，对其进行经验模态分解(EMD)，提取6层模态分量(IMF)进行Hilbert变换。实验结果显示，不论风机在空载、半载以及额定负载下，在匝间短路开始时，根据第4或者第5层IMF分量的瞬时幅值可以确定出匝间短路的収生，从而可以进行及时的维修，具有一定的工程应用价值。%When occurrence of doubly-fed wind generator stator winding slight inter-turn short circuit, the signal mutation caused by fault is very weak, the three-phase stator current changes very little, but the HHT analysis can obtain the time-frequency representation of the signal, it has obvious change when the fault occurs. So the fault signature can be found. The paper makes test in the dynamic model and simulation laboratory to obtain the current signal when the inter-turn short fault occurs. Empirical mode decomposition (EMD) is used, and to extract six Intrinsic modes to do their Hilbert-Huang transform. Experimental results show that instantaneous amplitude of the fourth or fifth Intrinsic mode determines if inter-turn short fault occurs or not, at no-load, half-load and rated load. It helps timely maintenance in practice, and has possibility in inter-turn short fault diagnosis.