Loading...

Structural Parameters Estimation Using Response Power Spectral Density

Gharechaei, Maryam | 2015

823 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47040 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Rahimzade Rofouei, Fayyaz; Esfandiari, Akbar
  7. Abstract:
  8. Damage and collapse in structural systems due to structural loads are considered as a threat to human lives. Therefore, using structural health monitoring methods in order to estimate structural damages and preventing structural collapse, have drawn a lot of attention. Due to the importance of considering random vibration, a nondestructive damage detection method using power spectral density function has presented in this study. This method is based on model updating and solving the sensitivity equations and it is also a model based method. The mentioned method can be classified as a vibration-based health monitoring method in frequency domain which includes dynamic excitation in some degrees of freedom and measuring power spectral density response of the structure utilizing sensors in several degrees of freedom. Using the proposed algorithm, it is feasible to detect damage, its location and severity by assessing the measured power spectral density function in the measuring points and a method to estimate the structural properties in unmeasured points. In this study, damage is considered to be linear and it is supposed as the reduction in elasticity modulus. By solving the sensitivity equations using least square method, damage can be estimated. After solving the equation in each step, stiffness matrix is calculated by the updated considered damage and damage estimation will be continued. A program in MATLAB compiler is used to assess different multi-damage scenarios in numerical models like a truss and a 2-D frame to validate the proposed algorithm. In order to consider the experimental errors in excitation and measurements, Mass modeling error of the intact structure and the errors in measuring the natural frequencies, random noise and errors are considered. The numerical results show the efficiency and accuracy of the proposed method in detecting the location and severity of damage
  9. Keywords:
  10. Least Squares Method ; Structural Health Monitoring ; Power Spectral Density (PSD)Analysis ; Vibration-Based Health Monitoring ; Structural Model Updating ; Sensitivity Equations

 Digital Object List

 Bookmark

No TOC