A Novel In Situ Wireless Monitoring of Bearing Cage Vibration and Skidding via Integrated Triboelectric and MEMS Sensing

ABSTRACT To address the limitations of traditional bearing condition monitoring, such as long vibration transmission paths and weak fault features easily corrupted by noise, this study proposes a novel wireless self‐sensing intelligent bearing (SV‐WSB) for in‐situ skidding and vibration detection. Integrated directly on the bearing cage with a Microelectromechanical Systems (MEMS) vibration sensor, triboelectric sensing unit, signal processing circuit, and Wi‐Fi module, the structure enables synchronous multi‐source signal acquisition and wireless transmission during bearing operation. The floating freestanding triboelectric module captures cage rotational characteristics without motion interference, while the MEMS sensor directly perceives cage vibration and transient responses. Experiments under steady‐state, acceleration‐deceleration, variable‐load and external impact conditions verify that triboelectric signals monitor cage speed and skidding ratio with only 0.2% deviation from high‐speed camera measurements. Vibration signals exhibit high sensitivity to transient impact excitation and unstable operating conditions. The influences of dielectric material and electrode gap on triboelectric output performance are systematically investigated. A Convolutional Neural Network (CNN)‐Transformer‐based multi‐source information fusion model achieves higher fault diagnosis accuracy than single‐signal methods. The proposed SV‐WSB provides a compact and effective solution for intelligent bearing condition monitoring and fault diagnosis.