Weighted Reconstruction and Improved Eigenclass Combination Method for the Detection of Bearing Faults

Table 2

Fault diagnosis procedure.

---

Diagnostic procedure

---

Step 1: decompose the original signal by TVFEMD to extract the IMFs component

Step 2: obtain the correlation-kurtosis index of each modal function and weight and reconstruct the signal

Step 3: use an FRFT filter for secondary noise reduction to improve the signal-to-noise ratio

Step 4: extract 18 feature indexes including time domain and entropy, and then, use the IDE for feature selection to form a feature set

Step 5: divide the data into the training set and the test set and combine them into the eigenclass for classification prediction

---