Progressive wear in rolling bearings involves coupled changes in surface morphology and vibration response, yet their stage-resolved evolution remains insufficiently quantified. This study develops a stage-resolved framework based on a 326 h rolling-contact wear experiment with 29 successive surface inspections. Three-dimensional raceway topographies were characterized using hierarchical descriptors of height distribution, spatial organization, and extreme-depression topology. Principal component analysis combined with piecewise-linear segmentation identified three morphology-resolved regimes corresponding to Running-in Wear, Steady-State Wear, and Severe Wear, with nominal transition times of approximately 95 h and 245 h. Moving-block bootstrap analysis indicated a gradual first transition and a more distinct onset of severe wear. Inspection-grouped ResNet50 analysis showed strong morphological separability among the three stages, with a pooled image-level accuracy of 90.42%. After controlling for operating time and multiple testing, surface roughness R q remained positively associated with vibration RMS ( r partial = 0.716 , q = 0.00132 ) and negatively associated with spectral entropy ( r partial = − 0.639 , q = 0.00906 ). Incorporating R q into a time-based RMS model reduced temporally blocked cross-validation RMSE by 27.8%. These results demonstrate that severe rolling-contact degradation is characterized by increasingly heterogeneous and localized surface morphology together with a stronger morphology–vibration association.
No takes yet. Share an insight, caveat, or question.
Yang et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: