• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2021, Vol. 57 ›› Issue (23): 160-168.doi: 10.3901/JME.2021.23.160

Previous Articles     Next Articles

Stochastic Resonance Induced by Asymmetric Potentials Enhanced Mechanical Repetitive Transient Extraction

QIAO Zijian1,2,3,4, SHU Xuedao1,3,4   

  1. 1. School of Mechanical Engineering and Mechanics, Ningbo University, Ningbo 315211;
    2. The State Key Laboratory of Mechanical Transmissions, Chongqing University, Chongqing 400044;
    3. Key Laboratory of Impact and Safety Engineering, Ministry of Education of China, Ningbo 315211;
    4. Zhejiang Provincial Key Laboratory of Part Rolling Technology, Ningbo 315211
  • Received:2020-05-01 Revised:2021-08-01 Online:2021-12-05 Published:2022-02-28

Abstract: Aiming to solve the issue that the mechanism of SR induced by well-depth or well-width asymmetry individually for enhancing repetitive transients has not been investigated, three kinds of asymmetric double-well potentials including well-depth asymmetric ones, well-width asymmetric ones and well-depth-width asymmetric ones are designed to propose a SR method induced by asymmetric potentials for enhancing mechanical repetitive transients. The proposed method utilizes the residence-time difference of Brownian particles between left and right potential wells to harvest noise energy for enhancing repetitive transients. Simulation and experiments of multi-stage gearboxes are performed to demonstrate its effectiveness. The results show that narrow and deep left potential well benefits to enhancing repetitive transients and the steepness of potential barrier would influence the SR for enhancing repetitive transients. In addition, the proposed method can diagnose incipient faults of multi-stage gearbox with an early crack on the tooth root of a planet gear and is superior to EMD. Even, the original signal's kurtosis index is increased by more than 80 times using the proposed method whose kurtosis index is 4 times of that of EMD.

Key words: incipient fault diagnosis, repetitive transient extraction, stochastic resonance, weak characteristic enhancement

CLC Number: