Inverse substructure method for model updating of structures

Abstract

Traditional model updating of large-scale structures is usually time-consuming because the global structural model needs to be repeatedly re-analyzed as a whole to match global measurements. This paper proposes a new substructural model updating method. The modal data measured on the global structure are disassembled to obtain the independent substructural dynamic flexibility matrices under force and displacement compatibility conditions. The method is extended to the case when the measurement is carried out at partial degrees-of-freedom of the structure. The extracted substructural flexibility matrices are then used as references for updating the corresponding substructural models. An orthogonal projector is employed on both the extracted substructural measurements and the substructural models to remove the rigid body modes of the free-free substructures. Compared with the traditional model updating at the global structure level, only the sub-models at the substructural level are re-analyzed in the proposed substructure-based model updating process, resulting in a rapid convergence of optimization.Moreover, only measurement on the local area corresponding to the concerned substructures is required, and those on other components can be avoided. The effectiveness and efficiency of the proposed substructuring method are verified through applications to a laboratory-tested frame structure and a large-scale 600 m tall Guangzhou New TV Tower. The present technique is referred to as the inverse substructuring model updating method as the measured global modal data are disassembled into the substructure level and then the updating is conducted on the substructures only. This differs from the substructuring model updating method previously This is the Pre-Published Version. 2 proposed by the authors, in which the model updating is still conducted in the global level and the numerical global modal data are assembled from those of substructures. That can be referred to as the forward substructuring model updating method.

Substructuring approach to the calculation of higher-order eigensensitivity

“…Thirdly, the number of parameters in each substructure is much smaller than that in the global structure. This improves the convergence of optimization process [23][24][25][26].…”

“…Substructuring technology is preferable for the analysis of large complex structures [18][19][20][21][22][23][24][25][26]. Firstly, it is possible to analyze each substructure independently, or even concurrently with parallel computing.…”

Structural Damage Detection Using Auto/Cross-Correlation Functions Under Multiple Unknown Excitations

“…The basic idea of frequency domain methods is that modal parameters (such as frequencies and mode shapes) are functions of physical parameters. The modal parameters are extracted via modal identification methods 1 , and the structural parameters are subsequently identified using model updating techniques 2,3 . The time domain methods use the measured time history responses directly for damage detection.…”

Construction of Stiffness and Flexibility for Substructure-Based Model Updating

“…The substructural solutions are then assembled to recover the solutions of the global structure by imposing constraints on the interfaces. On the other hand, the substructuring approach can be used in an inverse manner to disassemble the properties of the global structure to the substructure level by satisfying the constraints at the interfaces [15][16][17][18][19][20][21]. After eliminating the rigid body components, the independent substructures can be singled out to be used for the static analysis, 2 Mathematical Problems in Engineering dynamic analysis, nonlinear analysis, fatigue analysis, and so forth.…”

“…This paper addresses some frequently encountered difficulties associated with the analysis of the free-free substructures when the authors studied the substructuring methods in the previous research [5,[9][10][11][12][13][14][15][16]. A new full-rank stiffness matrix is proposed, which leads to a well-conditioned eigenequation.…”