基于ITD和LS - SVM的风力发电机组轴承故障诊断

为了更好地识别出复杂条件下风力风电机组主轴承的运行状态,提出了基于固有时间尺度分解(ITD)和最小二乘支持向量机(LS-SVM)的风电机组轴承故障诊断方法。该方法首先将调心滚子轴承振动信号分解成若干个固有旋转分量和一个趋势分量之和。然后,对前几个固有旋转分量的瞬时幅值进行频谱分析,找出频谱中外圈、内圈、滚动体故障特征频率处以及转动频率处的幅值,将其作为故障特征向量。最后,将故障特征向量输入LS-SVM来识别机组轴承的运行状态。实验结果表明,该方法可以快速、较准确地诊断出风力发电机组轴承故障。

Bearing fault diagnosis based on ITD and LS-SVM for wind turbine

In order to better identify the complex running conditions of main shaft bearings,the fault diagnosis based on ITD(Intrinsic Time-scale Decomposition) and LS-SVM(Least Square-Support Vector Machine) is proposed for wind turbine,which decomposes the bearing vibration signal into several proper rotation components and a trend component,analyzes the spectrum of instantaneous amplitude for the first few proper rotation components,finds the fault feature frequencies of outer race,inner race and roller,takes their amplitudes as the fault feature vectors,and inputs these fault feature vectors to LS-SVM to identify the operating conditions of bearings. Experimental results show that,the proposed method can quickly and more accurately diagnose the faults of wind turbine bearings.