Rolling element load and path analysis in electromechanically coupled digital twin modeling of rotating machinery

Abstract Establishing an accurate dynamic model for rolling bearings and analyzing their motion characteristics under varying operating conditions is crucial for identifying dynamic behaviors caused by bearing defects, which aids in effective fault diagnosis and prediction. Existing dynamic models often simplify bearing behavior by neglecting external electromechanical coupling effects from drive systems and assuming constant speeds. To address this gap, this paper develops a deep groove rolling bearing model within the framework of multi-body system dynamics, treating each component as an independent entity and considering their interactions. The stepper motor model, based on electromagnetic theory, is integrated with the bearing model to create a digital twin of the electromechanically coupled bearing test rig. Experimental validation confirms that the model accurately captures vibration waveforms, fault frequencies, and rolling element trajectories under varying defect sizes and speeds. Furthermore, the model can precisely simulate and predict the rolling element trajectory and contact load in the fault zone, offering a more accurate depiction of contact between bearing components. This study provides a theoretical foundation for industrial fault diagnosis and predictive maintenance strategies.