The Improved Dynamics Model of Faulty Bearings Considering Raceway Surface Fatigue Defect Morphology

Abstract A better understanding of the vibration characteristics of bearings with fatigue defects is of great significance for both remaining useful life assessment and fault diagnosis. This study investigates the vibration and impact responses that occur when rolling elements traverse surfaces containing fatigue defects. A Gaussian filter function-based modeling method for fatigue defect morphology is first proposed, in which cross-sectional slicing is applied to obtain the motion trajectory of the rolling elements as they traverse the defect. Subsequently, a nonlinear dynamic model incorporating the defect morphology is established. The derived motion trajectory is introduced into the model as a fault excitation function. During the modeling process, the finite size of rolling elements is considered, and an individual Lagrangian equation is formulated for each rolling element, thereby capturing in detail the dynamic behavior at defect locations and the variations in contact forces with both the inner and outer raceways. Finally, experimental validation is conducted. Results demonstrate that the proposed defect morphology modeling method achieves a Pearson correlation coefficient of up to 0.6 with the actual fatigue defect profile. Furthermore, the constructed multibody nonlinear dynamic model, which accounts for defect morphology, effectively reproduces the multiple impact phenomena observed in fatigue-defected bearings. It is also revealed that the impact timing of rolling elements passing defect positions is influenced by the external load.