• CN: 11-2187/TH
  • ISSN: 0577-6686

›› 2009, Vol. 45 ›› Issue (4): 166-171.

• Article • Previous Articles     Next Articles

Extracting Impact Feature of Machine Fault by Using Enhanced Filtering

SHAO Yimin;ZHOU Xiaojun;OU Jiafu;CHEN Zaigang;ZHOU Benxue   

  1. State Key Laboratory of Mechanical Transmission, Chongqing University China Automotive Engineering Research Institute
  • Published:2009-04-15

Abstract: The feature signal which can indicate whether the machine is in good condition or not is difficult to extract when the machine is operating under heavy noise environmental condition, specially early in the failure development. According to the properties of the mechanical noise and the impulse feature signal, a new method coupled by the evolutionary adaptive filtering and the wavelet denoising is proposed. The evolutionary adaptive filtering uses the cloning method and mating method to get the generations’ evolvement so as to realize the global optimization, while the wavelet denoising is realized by the reconstruction of the wavelet coefficient. Simulation and experiment results show that the signal to noise ratio are greatly improved and the algorithm has an excellent effect on processing the non-linear noise and extracting the early feature signal.

Key words: Enhanced filtering, Feature extracting, Machine fault

CLC Number: