Vibration-based structural damage detection and system identification using wavelet multiresolution analysis / Seyed Alireza Ravanfar

Ravanfar, Seyed Alireza (2017) Vibration-based structural damage detection and system identification using wavelet multiresolution analysis / Seyed Alireza Ravanfar. PhD thesis, University of Malaya.

[img] PDF (The Candidate's Agreement)
Restricted to Repository staff only

Download (1648Kb) | Request a copy
    [img]
    Preview
    PDF (Thesis (Ph.D.))
    Download (9Mb) | Preview

      Abstract

      During the service life of various civil, mechanical and aerospace structures, damage can nucleate, accumulate and propagate leading to out-of-service conditions which are dangerous and can sometimes collapse. Therefore, Structural Health Monitoring (SHM) is a crucial tool for identifying the presence and the evolution of possible damage. This thesis, investigates rigorously two paramount concerns of the SHM: damage detection and parametric system identification. The first proposed method is applied to detect crack damage in a structure. The location of the crack is identified by defining the damage index called relative wavelet packet entropy (RWPE). Then, the damage severities at the identified locations are assessed using genetic algorithm (GA), through defining a database to reveal the relationships between the energies obtained and damage severities. However, most of existing damage detection methods requires reference data which are not always available. Meanwhile, there has also been a pressing need for real-time monitoring to avoid sudden catastrophic disasters. Therefore, a new reference-free damage detection algorithm is proposed. The RWPE measurements of different sensor-to-sensor pairs are applied for defining the reference-free damage index (RDI) of each sensor location. To improve the proposed algorithm, GA was utilized to identify the best choice for ‘‘mother wavelet function” and “decomposition level” of the signals by means of the fundamental fitness function. This resulted in the high accuracy of the damage detection algorithm. The second proposed method seeks to identify damage in the structural parameters of linear and nonlinear systems. Initially, the connection coefficients for the scaling function of the proper selection of Daubechies wavelets are developed to derive the velocity and displacement from the measured acceleration responses. Then, the next step is to define the dominant components according to the relative energy distributions of wavelet packet transform (WPT) components of the acceleration responses, and transforming the equations of motion of the system in the time-domain to a reduced representation of the equations of motion based on the WPT. Finally, the least square error minimization method is implemented across the dominant components to determine the structural parameters of a linear system. Moreover, wavelet multiresolution analysis is applied to identify the tangent stiffness matrix and the hysteresis-restoring force of nonlinear structural systems without prior assumptions about the nonlinear characteristics of the systems. To demonstrate the validity and accuracy of the methods, numerical and experimental studies are conducted on a beam element and subsequently on a three-story building model. Results indicate that the wavelet-based damage detection method precisely identified the location and severity of damages even without reference data. In addition, the structural parameters of a system can be accurately estimated through the proposed system identification methods for both cases of linear and nonlinear conditions. Moreover, the accuracy and reliability of the proposed methods are investigated on various damage scenarios with different levels of severity, and noise levels.

      Item Type: Thesis (PhD)
      Additional Information: Thesis (Ph.D.) -– Faculty of Engineering, University of Malaya, 2017
      Uncontrolled Keywords: Wavelet multiresolution analysis; Reference data; Vibration-based structural damage detection; Linear and nonlinear conditions
      Subjects: T Technology > TA Engineering (General). Civil engineering (General)
      Divisions: Faculty of Engineering
      Depositing User: Mr Mohd Safri Tahir
      Date Deposited: 07 Jul 2017 15:57
      Last Modified: 07 Jul 2017 15:57
      URI: http://studentsrepo.um.edu.my/id/eprint/7435

      Actions (For repository staff only : Login required)

      View Item