利用随机共振的叶片裂纹微弱信息增强方法

Weak Fault Enhancement Method for Blade Crack by Using Stochastic Resonance

离心式压缩机叶片作为压缩机内最重要部件,长期承受周期性振动和流动诱使激励的作用。而叶片的故障将对压缩机的运行以及现场安全可靠性有严重的影响,因此如何有效地识别压缩机叶片裂纹早期故障显得尤为重要。由于叶片裂纹故障属于低频微弱故障,通常被调制到叶片通过频率处,但是故障频率难以识别,清晰度较低。首先在叶片通过频率处进行信号滤波,然后应用Woods-Saxon and Gaussian Potential随机共振模型对特征频率进行加强,从而得到叶片裂纹故障频率。通过在叶片裂纹附近安装压力脉动传感器,利用压力脉动信号对叶片裂纹信息进行监测。实现模拟叶片裂纹的信号测试,验证了WSG随机共振模型在叶片裂纹早期故障识别中的有效性以及可靠性。同时通过应变试验进行验证此方法的有效性。

For centrifugal compressor, blades as the most important part suffer periodic vibration and flow induced excitation mechanism. Blade failure has a serious impact on the operation of the compressor as the safety and reliability of the scene. Therefore, how to effective identify compressor blade crack fault is particularly important. As the signal of blade crack failure belongs to low frequency weak fault, the fault frequency is often modulated to the blade passing frequency. It is difficult to classify the characteristic frequeney. In this research, filtering at the blade passing frequency is firstly used for preprocessing. Then, the method of Woods-Saxon and Gaussian Potential stochastic resonance is applied to enhance the fault frequency demonstration. It is helpful for blade crack fault characteristic frequency determination. In this research, pressure pulsation （PP） sensors arranged in close vicinity to crack area are used to monitor the blade crack information. It can be concluded that WSGSR model is an effective tool for blade crack early fault detection on centrifugal compressor. As well, the strain experiment is carried out to verify the blade crack fault frequency.