A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Structural damage detection using nonlinear parameter identification with Tikhonov regularization
10.1002/stc.164.abs
A sensitivity‐based update of a finite element model is used to determine stiffness reduction factors from measured eigenfrequencies and mode shapes. This is generally an ill‐posed nonlinear problem that needs some regularization. Tikhonov regularization is combined with iterative linearization. Regularization has to be applied before linearization, otherwise the regularization effect depends strongly on the number of iterations. A nonlinear version of the generalized cross‐validation method is used to determine the optimal regularization parameter.
The theory is applied to a two‐dimensional, two‐storey frame. Numerical simulations with random errors are used to demonstrate the effect of regularization. The examples show that regularization can greatly improve the results and that generalized cross‐validation leads to selecting a good regularization parameter. The proposed method is also applied to experimental data obtained from a structure tested in the laboratory at the University of Sherbrooke. Results are in good agreement with experiments, although they depend somewhat on the weighting factors and initial values used for the iterative update. Copyright © 2006 John Wiley & Sons, Ltd.
Structural damage detection using nonlinear parameter identification with Tikhonov regularization
10.1002/stc.164.abs
A sensitivity‐based update of a finite element model is used to determine stiffness reduction factors from measured eigenfrequencies and mode shapes. This is generally an ill‐posed nonlinear problem that needs some regularization. Tikhonov regularization is combined with iterative linearization. Regularization has to be applied before linearization, otherwise the regularization effect depends strongly on the number of iterations. A nonlinear version of the generalized cross‐validation method is used to determine the optimal regularization parameter.
The theory is applied to a two‐dimensional, two‐storey frame. Numerical simulations with random errors are used to demonstrate the effect of regularization. The examples show that regularization can greatly improve the results and that generalized cross‐validation leads to selecting a good regularization parameter. The proposed method is also applied to experimental data obtained from a structure tested in the laboratory at the University of Sherbrooke. Results are in good agreement with experiments, although they depend somewhat on the weighting factors and initial values used for the iterative update. Copyright © 2006 John Wiley & Sons, Ltd.
Structural damage detection using nonlinear parameter identification with Tikhonov regularization
Weber, Benedikt (author) / Paultre, Patrick (author) / Proulx, Jean (author)
Structural Control and Health Monitoring ; 14 ; 406-427
2007-04-01
22 pages
Article (Journal)
Electronic Resource
English
Structural damage detection using nonlinear parameter identification with Tikhonov regularization
| Online Contents | 2007
Comparative studies on damage identification with Tikhonov regularization and sparse regularization
| Wiley | 2016
Modified Tikhonov regularization in model updating for damage identification
| British Library Online Contents | 2012
Spherical Tikhonov regularization wavelets in satellite gravity gradiometry with random noise
| Online Contents | 2001
Generalized Tikhonov Regularization Method for Large-Scale Linear Inverse Problems
| British Library Online Contents | 2013