Loading...

Evaluating two Model-Free data Interpertation Method for Measurments that are Influenced by Noise and Environmental Variations

Sedighi, Sasan | 2017

650 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49279 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Abedian, Ali
  7. Abstract:
  8. Interpreting data of infrastructures monitored to detect damage of a structural is one of the most challenging issues of structural health monitoring (SHM). Especially, when environmental factors such as temperature, wind and humidity affect the data and the situation become more complicated. In this study, two methods of analysis free model of moving principal components analysis (MPCA) and robust regression analysis (RRA) have recently been considered for monitoring infrastructures compared with a variety of methods of data interpretation and results are presented. In this study, the performance of different ways in different loading conditions and arrangement of sensors have been investigated. The results show that the function of method MPCA is highly affected by temperature, so in this research by altering destroyed place and changing the loading intensity that are affected by temperature period have been examined. One of the problems of method MPCA is to select moving window size. In real, it will be shown that moving window size is a function of pressure and temperature. Therefore, these impacts on gas transmission pipes commonly used have been examined to correctly estimate the size of the moving window. In this study, various scenarios are defined to detect damage and the results show that the method MPCA has a high damage detection feature.In this study, the MPCA is also investigated in the presence of improper data. The results show that for successive damages, MPCA method is reliable
  9. Keywords:
  10. Structural Health Monitoring ; Robust Regression ; Moving Window ; Damage Detection ; Temperature Variation ; Moving Principal Components Analysis (MPCA)

 Digital Object List

 Bookmark

No TOC