Advanced search options

Advanced Search Options 🞨

Browse by author name (“Author name starts with…”).

Find ETDs with:

in
/  
in
/  
in
/  
in

Written in Published in Earliest date Latest date

Sorted by

Results per page:

Sorted by: relevance · author · university · dateNew search

Language: English

You searched for subject:(Natural Excitation Technique). Showing records 1 – 2 of 2 total matches.

Search Limiters

Last 2 Years | English Only

No search limiters apply to these results.

▼ Search Limiters


University of Melbourne

1. Mahmood, S. M. Faisal. State-space model identification for modal-based damage detection.

Degree: 2012, University of Melbourne

A modal-based global method for identification, localization and quantification of damage of shear frame structures is presented in this thesis. The method involves identification of modal parameters from the vibration responses of the structure only (output-only) using a combination of Natural Excitation Technique (NExT) and Eigensystem Realization Algorithm (ERA), and determination of stiffness values using very few identified modes by employing a least squares solution of the eigenvalue problem. Finally, damage is indicated by comparing the stiffnesses of the (possibly) damaged structure with that of the undamaged structure. The implementation of NExT-ERA requires a reference channel and an order of the state-space model for modal identification, but several limitations of using a single reference channel and a particular order of the state-space model have been identified from the investigation of this study. To overcome the limitations, a new approach called Multi-Reference Based Mode Selection is proposed in this study. The approach makes use of all the available degree-of-freedoms (DOFs) as reference channels, but one at a time, to identify the best representative modes of the structure. An autonomous modal identification algorithm is also developed for fast online application of the approach. The stiffness values of a shear frame structure can be determined by employing the least squares solution of the eigenvalue problem. Although the solution can be obtained by using a single natural frequency and the corresponding mode shape, it has been found from this study that the errors in the determination of stiffness values are significant when just one mode is used in the calculation. At least two modes are required to determine the stiffness values with reasonable accuracy. Consequently, two modes were used in the determination of stiffness values using the least squares solution of the eigenvalue problem. The proposed damage detection methodology was implemented on the analytical Phase I and the experimental Phase II of the IASC-ASCE SHM benchmark problem. The negligible errors in the determination of modal parameters affirm that the methodology is robust in the presence of noise, but the methodology is found to be mildly sensitive to modeling errors, and because of that minor damage locations cannot be identified with certainty. Overall, from the investigation, it can be reported that the proposed methodology has an accuracy of around 90% in damage localization, and the determined severities of damages using it are very close to the actual ones.

Subjects/Keywords: damage detection; modal identification; Natural Excitation Technique; NExT; Eigensystem Realization Algorithm; ERA; Multi-Reference Based Mode Selection approach; benchmark problem

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Mahmood, S. M. F. (2012). State-space model identification for modal-based damage detection. (Masters Thesis). University of Melbourne. Retrieved from http://hdl.handle.net/11343/37298

Chicago Manual of Style (16th Edition):

Mahmood, S M Faisal. “State-space model identification for modal-based damage detection.” 2012. Masters Thesis, University of Melbourne. Accessed March 03, 2021. http://hdl.handle.net/11343/37298.

MLA Handbook (7th Edition):

Mahmood, S M Faisal. “State-space model identification for modal-based damage detection.” 2012. Web. 03 Mar 2021.

Vancouver:

Mahmood SMF. State-space model identification for modal-based damage detection. [Internet] [Masters thesis]. University of Melbourne; 2012. [cited 2021 Mar 03]. Available from: http://hdl.handle.net/11343/37298.

Council of Science Editors:

Mahmood SMF. State-space model identification for modal-based damage detection. [Masters Thesis]. University of Melbourne; 2012. Available from: http://hdl.handle.net/11343/37298


University of New South Wales

2. Chowdhury, Mohammad Shakar. Application of System Identification using ERA/NExT for Damage Assessment in Composite Laminates.

Degree: Engineering & Information Technology, 2017, University of New South Wales

Polymer matrix composites are widely used in aircraft structures due to their superior specific stiffness and strength. However, laminated composites are susceptible to delaminations due to the low interlaminar fracture toughness of the matrix. Traditional NDI techniques for damage detection involve significant costs due to the requirement for the grounding of aircraft, shutdown time, and labour involved in inspections. Vibration monitoring for damage assessment in aircraft has several advantages such as being applicable in situ, not requiring external excitation and the potential for application as a continuous structural health monitoring system. Frequency monitoring is employed in modern aircraft for engine diagnostics and identification of major faults; however, when applied to structural components, monitoring frequency shifts only identifies the presence of damage but not its exact location and severity. Eigensystem Realization Algorithm (ERA) has been proposed for application to many structural components but not for assessment of delamination in composites.This thesis proposes the application of ERA coupled with Natural Excitation Technique (NExT) for delamination assessment in laminated composite structures, including stiffened composite panels. NExT provides the advantages that any type of excitation of the structure can be employed and the location of the excitation point need not be known. The algorithm for delamination assessment using ERA coupled with NExT was developed using Matlab code. The proposed method is first investigated and validated using transient vibration response generated by finite element simulation of composite laminates with delaminations. The simulated response is fed into the ERA/NExT algorithm to identify changes in the system parameters to estimate the location and reductions in bending stiffness in the damaged areas. The use of different types of excitation modes and excitations provided at different locations have been examined. Sensitivity analyses have been performed conducted study the influence of spacing of measurement points on the accuracy of delamination size and severity estimated using responses by numerical modelling of composite beams and laminated plates. The proposed method is validated experimentally with vibration measurements conducted on laminated carbon/epoxy beams, plates and stiffened panels with embedded delaminations and estimating the location, size and severity of the delaminations in the samples from the measured dynamic responses. Advisors/Committee Members: Shankar, Krishna, Engineering & Information Technology, UNSW Canberra, UNSW, Morozov, Evgeny, Engineering & Information Technology, UNSW Canberra, UNSW.

Subjects/Keywords: Eigensystem Realisation Algorithm; Vibration; Structural Health Monitoring; Natural Excitation Technique; Damage Assessment; Carbon Fiber Reinforced Polymer

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Chowdhury, M. S. (2017). Application of System Identification using ERA/NExT for Damage Assessment in Composite Laminates. (Doctoral Dissertation). University of New South Wales. Retrieved from http://handle.unsw.edu.au/1959.4/57569 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:44102/SOURCE02?view=true

Chicago Manual of Style (16th Edition):

Chowdhury, Mohammad Shakar. “Application of System Identification using ERA/NExT for Damage Assessment in Composite Laminates.” 2017. Doctoral Dissertation, University of New South Wales. Accessed March 03, 2021. http://handle.unsw.edu.au/1959.4/57569 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:44102/SOURCE02?view=true.

MLA Handbook (7th Edition):

Chowdhury, Mohammad Shakar. “Application of System Identification using ERA/NExT for Damage Assessment in Composite Laminates.” 2017. Web. 03 Mar 2021.

Vancouver:

Chowdhury MS. Application of System Identification using ERA/NExT for Damage Assessment in Composite Laminates. [Internet] [Doctoral dissertation]. University of New South Wales; 2017. [cited 2021 Mar 03]. Available from: http://handle.unsw.edu.au/1959.4/57569 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:44102/SOURCE02?view=true.

Council of Science Editors:

Chowdhury MS. Application of System Identification using ERA/NExT for Damage Assessment in Composite Laminates. [Doctoral Dissertation]. University of New South Wales; 2017. Available from: http://handle.unsw.edu.au/1959.4/57569 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:44102/SOURCE02?view=true

.