Order-Based Identification of Bearing Defects under Variable Speed Condition

Condition monitoring of rotating machinery plays an important role in reducing catastrophic failures and production losses the 4.0 Industry. Vibration analysis has proven to be effective diagnosing machine failures. However, identifying bearing defects based on vibration remains a difficult task, especially non-stationary operation conditions. This work aims automate process under variable operating speeds. Based order technique, three frequency domain features: Spectrum peak Ratio Outer (SPRO), Inner (SPRI), Rolling element (SPRR) are updated perform with signals. The features extracted from data real ball system. They retained build predictive multi-kernel support vector (MSVM) classification model. Therefore, effectiveness proposed is evaluated performance constructed classifier. deployed have their bearing: outer race, inner ball, combined speed