Loading...

Determination of Ball-Bearings Failure Thresholds by Using a Data-Driven Method

Feizhoseini, Mir Sajjad | 2021

693 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 53689 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Behzad, Mehdi
  7. Abstract:
  8. One of the challenges in predicting the remaining useful life (RUL) of rolling element bearings (REBs) is determining a proper failure threshold (FT). In the literature, the FT is usually assumed to be a constant value of an extracted feature from the vibration signals. In this research, a method was proposed to define the FT of REBs, and the effect of different parameters on the FT was also studied. First, to determine the effective parameters, the variance analysis was applied to the seeded fault dataset of REBs. Afterward, the method based on the principal component analysis (PCA) and the copula models was proposed to estimate the FT. In this method, a proper feature for the FT and a feature as a degradation indicator were extracted by applying PCA to the run-to-failure data, including the accelerated-life dataset and the industrial dataset. The relationship between this degradation indicator and the selected feature was represented through a joint probability distribution using statistical copula models. Finally, the FT was proposed as a probability distribution based on the fluctuation increase in the vibration trend. The results of the seeded fault dataset showed that the size and the type of defect, the velocity, and the radial force are the critical factors involved in the FT. Moreover, applying the proposed method to the accelerated-life dataset led to various FTs for the different failure modes that occurred. It is shown that a higher FT can be assumed for inner race degradation than for rolling element degradation. This could help extend the lives of REBs regarding the degrading elements. A dataset for an industrial machine was also analyzed, and it is shown that the proposed model estimated a reasonable and proper FT in an actual case study
  9. Keywords:
  10. Remaining Useful Life ; Variance Analysis ; Failure Modes ; Principal Component Analysis (PCA) ; Copula Functions ; Failure Threshold ; Rolling Bearing

 Digital Object List

 Bookmark

No TOC