Earticle

초록

영어

The vibration signal is one of the significant signals that reflects the fault. In allusion to the shortcomings of traditional signal analysis method in the high-frequency and non-stationary signal analysis, the wavelet theory and approximate entropy algorithm are introduced into the signal analysis in order to propose a new vibration signal analysis (WTAEAVSA) method in this paper. In the proposed WTAEAVSA method, the wavelet transform technology is used to reduce the noise and decompose the low and high frequency vibration signal in order to obtain the signal characteristics of different frequency bands. Then the approximate entropy algorithm is used to determine the complexity and irregular degree of vibration signal in the different scale and different frequency band, so as the non-stationary characteristics of vibration signal are extracted. At last, some simulated signals with time-domain and frequency-domain from the normal signal are used to test the characteristics of the proposed WTAEAVSA method. The simulation results show that the proposed WTAEAVSA method can extract the characteristic vector from vibration signal, visually and sharply reflect the changes of the mechanical states.

목차

Abstract
 1. Introduction
 2. Wavelet Theory
  2.1. Wavelet transform
  2.2. Wavelet Packet Transform
 3. Approximate Entropy Algorithm
 4. Feature Extraction Based On Wavelet Theory and Approximate Entropy Algorithm
 5. Experiment Result and Analysis
 6. Conclusion
 Acknowledgments
 References

키워드

저자

• Xiaoyong Yu [ School of Information Engineering, Suzhou University, Suzhou, 234000 China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Future Generation Communication and Networking

• 간기

  격월간

• pISSN

  2233-7857

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605