Loading...

Application of Different Signal Processing in Rolling Element Bearing Fault Diagnosis

Mohsenpour Ghazimahale, Ali | 2010

457 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 40636 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Behzad, Mehdi; Mehdigholi, Hamid
  7. Abstract:
  8. Rolling bearings are important parts in rotating machinery. Unsuitable practice of this bearing causes unsuitable output in this collection. In some machines like helicopters, sudden faults of this rolling bearing cause body dangers. According to this point, diagnose of bearing fault is an important case in condition monitoring domain. There are many ways for diagnose of fault in bearing which most of them are important subject in signal processing. In this project first of all, the most important and applicable signal processing methods, which are useful in diagnose of fault in bearings are introduced then three of these methods are chosen and applied on time signals of experimental models, and the results will be discussed. These 3 methods include: Envelope analysis, Bispectrum analysis and Cyclostationary
    In experiments which have been done in 2 sample rating after using the mentioned methods, it is observed that bispectrum analysis is sensitive to high sample rating, because the signal is converted to nonstationary signal. This will cause high fault in diagnosing but for nonlinear systems, and also for diagnosing phase information between frequencies is suitable. Therefore this method is useful for stationary signal. For other two methods, the results are to some extent similar. but according to simple process of envelope analysis, the use of this method is recommented. It is important to say that in time signals which are complicated, using of cyclostationary is suggested, and this can be an introduction for doing envelope analysis
  9. Keywords:
  10. Envelope Analysis ; Quadratic Phase Coupling ; Rolling Bearing ; Bispectrum ; Cyclostationary

 Digital Object List

 Bookmark

No TOC