Back to results list
Please use this identifier to cite or link to this item:
|Title:||Incomplete dynamic measurement in structural damage assessment||Authors:||Mak, Pui-shan||Keywords:||Structural engineering
Hong Kong Polytechnic University -- Dissertations
|Issue Date:||2001||Publisher:||The Hong Kong Polytechnic University||Abstract:||The structural damage assessment strategy adopted in this thesis is to localize the damage sites first using incomplete measured mode shapes, and to quantify the structural damage extent later using measured frequencies which are less contaminated by measurement noise and with better accuracy than mode shapes.
Modal strain energy change (MSEC) has been proved to be more sensitive to structural damage than natural frequencies and mode shapes. The parameter MSEC is used in the correlation matching of the Multiple Damage Location Assurance Criterion (MDLAC) to localize the damage. The sensitivity of MSEC is a function of the analytical property matrices, natural frequencies and mode shapes of the original structure. The incomplete measured mode shapes before and after damage and the analytical stiffness matrix are required to calculate the experimental MSEC which are then normalized. The potential damage sites are identified as those elements with higher MDLAC values. The MDLAC values on the different combinations of the potential damage sites are further computed. The final damage sites are located corresponding to the combination with the largest MDLAC value. Simulation studies of the damage localization method are based on a three-storey plane frame.
After localization of the damage sites, the damage extent of the final damage sites can be determined using the measured natural frequencies. A structural damage quantification method is developed using only the first order sensitivity terms in which a linear sequential filtering technique is used to relate the analytical and experimental sets of elgenvalues including the measurement noise matrix. An unbiased minimum variance error estimation method is used for an optimal solution. Weight linear least squares method is used to solve the identification equation. Simulation studies on the damage quantification method are conducted on the same three-storey plane frame.
The accuracy of identification using only the first order terms will be reduced when dealing with large damage, and the damage quantification method including the second order terms in the Taylor's expansion is proposed. Simulation studies on the three-storey plane frame are again performed.
In these two damage quantification methods, the formulation has the capability to estimate a larger set of eigenvalues from a smaller set of measured eigenvalues. The effectiveness of the eigenvalue expansion is also studied using the three-storey plane frame.
After simulation studies, a dynamic test is carried out in the laboratory on a five-storey steel plane frame to further verify the damage localization method and the damage quantification methods using first order and second order analysis. A method to select the modes for the identification is proposed.
|Description:||xxii, 165 leaves : ill. ; 30 cm.
PolyU Library Call No.: [THS] LG51 .H577M CSE 2001 Mak
|URI:||http://hdl.handle.net/10397/1002||Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b1600730x_link.htm||For PolyU Users||167 B||HTML||View/Open|
|b1600730x_ir.pdf||For All Users (Non-printable)||3.51 MB||Adobe PDF||View/Open|
Citations as of Jun 18, 2018
Citations as of Jun 18, 2018
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.