Loading...

Implementation of RUL Estimation Approaches on Software Platform, Concentrated on Confidence Level Determination in Rolling Element Bearings

Mirfarah, Motahareh | 2019

442 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 52648 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Behzad, Mehdi
  7. Abstract:
  8. Estimating the remaining useful life (RUL) of critical assets is an essential task in system health management. It also has a crucial role in the optimization of maintenance scheduling and decision making about the future of the asset. On the other hand, since the rolling element bearings are widely used in the rotating machinery, estimation of their RUL is considered as remarkable progress in improving the reliability of the whole system. In this way, various models have been introduced to predict the RUL of rolling element bearings, which their proper functionality is affected by the underlying assumptions in the model structure. Because of the presence of several uncertainties and complexities in the rolling element bearings degradation process, relying on the output prediction of one model in these components is not reliable and could result in misleading predictions. In the present project, it is proposed to construct an integrated prognostics framework by the fusion of models output predictions through Dempster-Shaefer theory (DST). In this way, the prediction of rolling element bearings RUL is performed by the implementation of three practical and useful models, including feed-forward neural network (FFNN), extended Kalman filter (EKF), and particle filter (PF). Then, the output predictions of these models are fused through the proposed DST framework, so that the confidence level and the robustness of prediction results will be improved. In order to investigate the effectiveness of the proposed DST framework, the experimental rolling element bearing datasets are utilized. Examination of the results illustrated that the point predictions of the proposed framework are more accurate, and its interval predictions have more reliability and quality in comparision to using an individual model. Therefore, the proposed fusion framework can be considered as an effective tool in making logical decisions about the future of the system
  9. Keywords:
  10. Remaining Useful Life ; Data Fusion ; Dempster-Shafer Theory ; Confidence Level ; Rolling Bearing ; Forecasting

 Digital Object List

 Bookmark

No TOC