基于复调制滤波与ESPRIT的异步电动机转子断条故障检测新方法

摘要

转子断条故障难以检测的原因在于工频分量与边频分量在频率上相隔较近且边频信号微弱,这一影响在低转差率时尤为严重。这种情况下,具有高频率分辨率的旋转不变信号参数估计技术（ estimation of signal pa-rameters via rotational invariance technique,ESPRIT)的估计效果也失去意义。因此,引入复调制技术,将工频分量搬移至0 Hz并予以滤除。同时,引入hilbert变换构成解析信号,利用其单边频谱性解决负频率分量问题。进而,提出了基于复调制滤波与ESPRIT的异步电动机转子断条故障检测新方法。仿真与实验验证了此方法的有效性。%Broken rotor bar fault is difficult to detect because of the fact that the power frequency component is so close to the side frequency component and the side frequency component is too weak, which is particularly serious in condition of low slip. In this case, even ESPRIT is ineffective. Therefore, a complex modulation technique is intro-duced to remove the power frequency component by shifting it to 0 Hz in frequency domain. Meanwhile, Hilbert trans-form is adoptedto solve the negative frequency problem by its one-sided spectrum characteristic. Finally, a new BRB detection method is presented by combining complex modulation techniques and ESPRIT seamlessly. Simulating re-sults on an induction motor verify the effectness.