一种优化轴速提取方法以及在风机故障检测中的应用

Optimal shaft speed extraction and its application in wind turbine condition monitoring

在风机故障监测中,同步采集和同步分析技术能够有效地减缓风机变速旋转导致的频谱能量弥散现象,使损伤特征频率能量集中,从而易于实现故障的精准检测。然而由于现有的风机故障检测系统通常并未配有速度传感器,因此从风机振动信号中提取瞬时轴相位是实现同步采集与分析技术的前提和关键。现有的相位解调轴速提取方法往往由于带宽的选择问题导致难以得到较为精确的相位信息,提出了一种优化轴速提取方法(optimal shaft speed extraction,OSSE),该方法基于代价函数脊线追踪(one-step cost function,OSCF)、最优带宽搜索及基于解析信号的相位解调方法对风机轴速进行提取。仿真结果表明,该方法能够获得精确的相位信息,使用该相位信息对原始振动信号同步重采集,可以得到清晰的阶次谱。利用某风场实际采集的风机振动数据进行验证,所得结论与仿真结果一致,证明了该方法的有效性和准确性。

In wind turbine condition monitoring,application of synchronous resampling and synchronous analysis techniques can greatly alleviate the spectral energy dispersion phenomenon caused by the variable speed operations,so that it is easier to realize damage feature extraction.However,the existing wind turbine condition monitoring system is usually not equipped with the speed sensor.Therefore,extracting the shaft instantaneous phase from the measured vibration responses is the key to realize synchronous resampling and analysis techniques.It is difficult to use the existing speed extraction methods with phase demodulation to obtain accurate phase information due to the problem of bandwidth selection and other restrictions.In this paper,an optimal shaft speed extraction(OSSE)method was proposed,based on the combination of one-step cost function(OSCF),optimal bandwidth searching and Hilbert phase demodulation,to extract the shaft instantaneous speed of wind turbine.The simulation results show that the proposed method can obtain accurate phase information.Using this phase signal to synchronously resample the vibration signal and carry out the spectrum analysis,a clearer order spectrum can be obtained.Finally,the vibration data from a real wind turbine operation were used to verify the proposed method.The analysis results derive the similar conclusions as those from simulations.The effectiveness and accuracy of the proposed method are demonstrated.