Abstract: Under the multi-source coupling strong noise interference, it is often difficult to extract the early fault characteristic signals of rolling bearings quickly and accurately. Aiming at the problems of weak anti-noise ability and low computational efficiency in existing research, this paper proposes a method for extracting early fault features of rolling bearings based on simulation-driven Ramanujan period transform. First, based on the prior knowledge of the rolling bearing to be analyzed, a dynamic model of the fault simulation is constructed and the steady-state response of the simulation is obtained. Secondly, the simulation signal is transformed by Ramanujan period to obtain transform coefficients, and the transform position of fault feature information is obtained based on the transform coefficients. Finally, based on the fault feature information obtained from the simulated response signal, the position is transformed, and the Ramanujan period transformation is performed on the actual monitoring signal to separate the early fault features. The proposed method is validated with the IMS bearing dataset, and the results show that the proposed method can efficiently and accurately extract the early fault features of rolling bearings under strong background noise.

Key words: rolling bearing, simulation model, Ramanujan periodic transform, early fault feature extraction