Abstract: With the increasing speed and operational density of trains, the safety margin of wheelsets is progressively diminishing. As a critical rotating and load-bearing component, the health of the double-row tapered roller bearing directly governs the operational safety, reliability, and life-cycle cost of the entire vehicle. To overcome the limitations of existing online monitoring systems, such as their reliance on external power and restricted sensor placement, this study proposes a Triboelectric Nanogenerator (TENG)-based in-situ sensing module (TISM). This module is designed to be coaxially mirror-symmetric with the cage and ultra-compactly integrated within the bearing. This configuration enables the self-powered, synchronous acquisition of relative motion data between the cage and rollers on both sides of the double-row bearing. Furthermore, the study establishes a qualitative mapping between the TENG output signal and key operational parameters—including rotational speed, load, and lubrication status—providing a solid foundation for real-time bearing condition identification. Experimental results demonstrate that the monitoring scheme possesses high sensitivity and self-powering capabilities. It also exhibits excellent robustness in terms of material selection and design clearance，providing crucial multi-physics parameter support for the construction of a digital twin for the train wheelset bearing. Validation on a railway vehicle lateral dynamics test rig confirms the TISM’s excellent monitoring effectiveness and self-sensing capabilities under complex variable-speed conditions. Through time-frequency analysis of the acquired signals, real-time perception of the bearing's operating state and precise detection of early fault signatures，such as cage slip, can be achieved. This research provides a new technical paradigm for the intelligent design and self-powered monitoring of rolling bearings, showcases a cyber-physical interaction model for a component-level digital twin，and has the potential to be extended to the digital twin framework of the entire train system.

Key words: triboelectric nanogenerator, train wheelset, smart bearing, self-powered sensing, digital twin