发电机转子绕组匝间短路故障诊断研究

Research on the Fault Diagnosis for Winding Inter-turn Short Circuit in Rotor of Power Generator

针对同步发电机转子匝间短路故障诊断特征提取困难、信号受噪声干扰较大、常规的故障表征量对轻微故障不敏感等问题,提出一种新型的转子绕组匝间短路故障诊断方法。首先,以定子电流信号为研究对象,分析同步发电机转子匝间短路后励磁磁场谐波特性;其次,提出小波包-关联维数和维数距离的概念,即利用小波包把信号分解到足够细的层次,引用分形理论计算各层细节信号的关联维数作为故障表征量;最后,根据不同状态下的维数距离进行故障诊断和故障程度判断。仿真结果表明,该方法可靠实用,为故障诊断提供了理论指导。

In accordance with the deficiencies exist in fault diagnosis of winding inter-turn short circuit in rotor of synchronous generator, e. g. , fault features are difficult to be extracted, the signal is influenced by noises, the amount of conventional fault characterization is insensitive to minor fault, etc. , the new type of fault diagnostic method is proposed. Firstly, with the stator current as the research object, the harmonic characteristics of excitation magnetic field after winding inter-turn short circuit are analyzed; then the concept of wavelet packet correlation dimension and dimension distance is proposed, i.e., the signal is hierarchically decomposed to fine enough by adopting wavelet packet decomposition, the correlation dimension of each hierarchical minutiae signal is calculated by adopting fractal theory and used as the fault characterization ; finally the fault diagnosis is conducted and fault degree is judged in accordance with the calculation of dimension distance under different states. The result of simulation shows that this method is reliable and practical; it provides theoretical guidance for fault diagnosis.