Based on SSA and MNAD Rolling Bearing Fault Feature Extraction Method
Anwen Tan, Yuanjun Dai · 2024
Aiming at the problem that the periodic transient shock is not obvious at the early stage of rolling bearing faults under background noise, this paper proposes adaptive Minimum noise amplitude deconvolution (MNAD) fault feature extraction algorithm. Firstly, Sparrow Search Algorithm (SSA) is used to optimize the fault frequency FCF and filter length L in MNAD to solve the problem of MNAD instability, and then, the Envelope Periodic Pulsing Factor (EPPF) index is proposed as the SSA adaptability function to iteratively search for the optimal solutions for FCF and L. Finally, the optimized filtered signal obtained by applying MNAD is analyzed by envelope demodulation to extract the fault signal accurately. Through the analysis of experimental data, it is shown that the method can effectively enhance the periodic transient shock characteristics of vibration signals.