Synchronous monitoring of rotational frequencies in inter-shaft bearing components via weak magnetic signals and adaptive ridge extraction

Abnormal cage slip often leads to the failure of inter-shaft bearings, which is used as a key indicator in bearing health management. It is calculated according to the differences in the rotational frequencies among the bearing components. However, the concurrent rotation of the inner and outer rings in inter-shaft bearings could introduce considerable complexity, and the traditional cage slip monitoring techniques may not perform well in inter-shaft bearing systems. To address this issue, this study proposes a non-contact method for synchronous monitoring of the rotational frequencies of inter-shaft bearing components based on weak magnetic fields, which also combines an adaptive ridge extraction algorithm. By capturing subtle variations in magnetic signals during bearing operation, the proposed method enables simultaneous monitoring of the rotational frequencies of the inner race, outer race, and cage without requiring any physical modification of the bearing structure. Only a single non-contact probe is required in our method, and our compact system shows strong adaptability to varying operating environment. The adaptive ridge extraction algorithm dynamically adjusts the penalty function based on the signal characteristics, which can balance frequency peak identification with trajectory continuity and effectively suppress the harmonic and noise interference generated by rolling elements. Under variable-frequency conditions, the proposed method can significantly enhance the accuracy and robustness of frequency extraction for each component. The experimental results prove the reliability and practicality of our method, and our method provides a promising nondestructive solution for online monitoring and failure analysis of inter-shaft bearings.