本研究利用希爾伯特-黃轉換針對旋轉機械的軸承元件於變轉速下，軸承發生外圈損壞、內圈損壞、滾子損壞和複合故障等情形進行故障診斷。首先，利用階次追蹤方法將轉速變動因子抽離，使非穩態的訊號轉換成穩態的角度域訊號，再提取角度域訊號之包絡線透過希爾伯特-黃轉換進行分析，於分解出來的固有模態函數和希爾伯特邊際譜探討軸承不同的損壞特徵，並經由振幅正規化後使得故障特徵不會隨著變轉速影響，最後，以支持向量機進行單一故障軸承的故障診斷，且利用此分類器對複合故障軸承進行故障診斷。 In this study, Hilbert-Huang transform (HHT) is utilized for diagnosing the roller bearing faults, such as outer race defect, inner race defect, roller defect and multi-fault, under variable rotation speed. The vibration signals are first measure through the order tracking technique, so that the vibration signals are detected with identical angle increment and thus the vibration signals are stationary without the factor of variable shaft rotation speed. The envelope signals of the measurements are analyzed by Hilbert-Huang transform approach. The features of the faulted bearings are extracted by investigating the intrinsic mode functions (IMFs) as well as the marginal Hilbert spectra. The extracted features of the faulted bearing are then re-scaled through the amplitude normalization, so that the vibration energy are not affected by the variable rotation speed. Finally, the support vector machine is employed to identify the individual defect of bearing. The same SVM structure is also used to diagnose the occurrence of multi-fault in bearings.
