基于阶次窄带包络分析的直驱风机轴承微弱故障诊断

Weak Fault Diagnosis Method for Direct-drive Wind Turbine Rolling Bearings Based on Order Narrow Band Envelope Analysis

针对变转速工况下直驱风电机组滚动轴承早期微弱故障难以诊断的技术难题,提出一种基于阶次窄带包络的滚动轴承早期故障诊断新方法。首先,求出变转速下直驱风电机组滚动轴承故障信号时频谱,并设定等效转频。再利用时频自动搜峰算法搜索出等效转频对应阶次时频峰值。进一步,采用最小二乘法拟合得到转速信号变化曲线,通过该转速曲线计算出重采样鉴相时标,并对原信号进行等角域重采样。最后,对采样后角域信号进行窄带包络分析得到含滚动轴承故障信号的阶次窄带包络谱,该阶次窄带包络谱可用于滚动轴承故障诊断。该方法同时克服了变转速影响和低频调制微弱信号难以提取的问题,为滚动轴承故障诊断提供新方法。直驱风机滚动轴承应用实例验证了方法的有效性。

A new method based on order narrow-band envelope （ONBE） is proposed to diagnose the early weak faultof direct-drive wind turbine rolling bearings. Firstly, the time spectrum of the fault signal of the direct-drive wind turbinerolling bearing under variable speed conditions is obtained, and the equivalent frequency is set. Then, the time- frequencyautomatic peak- searching algorithm is employed to search for the time and frequency peak values corresponding to theequivalent rotation frequency. Furthermore, the curve of speed signal is obtained by least squares fitting, resampling phasediscrimination labels are calculated based on the speed curve, and the original signals are resampled in an angle domain.Finally, the ONBE spectrum of the rolling bearing fault signal is obtained which can be used to diagnose the rolling bearingfaults. This method can eliminate the effect of the change of speed and can overcome the difficulty of extracting the lowfrequency weak signals. This method provides a new way for fault diagnosis of rolling bearings. The application example indirect-drive wind rolling bearings verifies the effectiveness of this method.