Non-Stationary Vibratory Signatures Bearing Fault Detection Using Alternative Novel Kurtosis-based Statistical Analysis

Vibration signature-based analysis to detect and diagnose is the commonly used technique in monitoring of rotating machinery. Reliable features will determine efficacy diagnosis prognosis results field machine condition monitoring. This study intends produce a reliable set signal through an alternative statistical characteristic before available relevant prediction methods. Given above advantage Kurtosis, newly formed feature extraction adapted extract single coefficient out EMD-based pre-processing vibration data for bearing fault detection Each IMFs analyzed using Z-rotation method coefficient. Afterwards, Z-rot coefficients, RZ are presented on base specification defect vibratory observe which IMF has highest correlation over given. Throughout studies, shows some significant non-linearity measured impact. For that reason, effectively determined strong existed components fault. It corresponds first inner race rolling ball specified with R2 = 0.9653 (1750 rpm) 0.9518 (1772 rpm), respectively. Whereas, 4th decomposition outer scored 0.8865 rpm). Meanwhile, average R-squared score between throughout 0.8915. Thus, it can be utilized findings conditions.