TY - JOUR
T1 - Structural Response Reconstruction Based on Experimental Approximation and Model Reduction
AU - Yang, Jinsong
AU - Liu, Yidan
AU - Lu, Xuandong
AU - Wang, Tiantian
AU - Xie, Jingsong
AU - He, Jingjing
N1 - Publisher Copyright:
Copyright © 2026 Jinsong Yang et al. Structural Control and Health Monitoring published by John Wiley & Sons Ltd.
PY - 2026
Y1 - 2026
N2 - Model-driven response reconstruction is widely applied to supplement inaccessible dynamic information in mechanical structural health monitoring. However, increasing complexity of structures with considerable degrees of freedom and nonshareability of substructures based on security issues make it inefficient, even impossible, to perform response reconstruction. To overcome these difficulties, a reconstruction strategy is presented by innovatively incorporating experimental approximation and model reduction. As the baseline of response reconstruction, the complete structural FEM is integrated by the defined FEM-based substructures (FSs) and experiment-based substructures (ESs). On one hand, FS aims to deal with low reconstruction efficiency and is transformed from the accessible substructure by the Crag–Bampton method that eliminates redundant matrix data when ensuring accuracy. On the other hand, ES is established to improve the condition of modeling difficulty that inaccessible FEM of local components is replaced by the approximate model, whose vibration characteristics are described through removing a predesigned fixture using experimental coupling measurement. The complete model is established by assembling the above substructures at the desired interface. Response reconstruction referred to the area of ES is realized through transferring frequency response functions among substructures. Otherwise, reconstruction could be directly accomplished by Eigen solution and inverse modal transformation. Response reconstruction cases of a numerical beam structure and an experimental flange-tube assembly are applied to verify superiority and accuracy of the proposed method.
AB - Model-driven response reconstruction is widely applied to supplement inaccessible dynamic information in mechanical structural health monitoring. However, increasing complexity of structures with considerable degrees of freedom and nonshareability of substructures based on security issues make it inefficient, even impossible, to perform response reconstruction. To overcome these difficulties, a reconstruction strategy is presented by innovatively incorporating experimental approximation and model reduction. As the baseline of response reconstruction, the complete structural FEM is integrated by the defined FEM-based substructures (FSs) and experiment-based substructures (ESs). On one hand, FS aims to deal with low reconstruction efficiency and is transformed from the accessible substructure by the Crag–Bampton method that eliminates redundant matrix data when ensuring accuracy. On the other hand, ES is established to improve the condition of modeling difficulty that inaccessible FEM of local components is replaced by the approximate model, whose vibration characteristics are described through removing a predesigned fixture using experimental coupling measurement. The complete model is established by assembling the above substructures at the desired interface. Response reconstruction referred to the area of ES is realized through transferring frequency response functions among substructures. Otherwise, reconstruction could be directly accomplished by Eigen solution and inverse modal transformation. Response reconstruction cases of a numerical beam structure and an experimental flange-tube assembly are applied to verify superiority and accuracy of the proposed method.
KW - complex mechanical structures
KW - experimental approximation
KW - model reduction
KW - response reconstruction
UR - https://www.scopus.com/pages/publications/105033073888
U2 - 10.1155/stc/5553171
DO - 10.1155/stc/5553171
M3 - 文章
AN - SCOPUS:105033073888
SN - 1545-2255
VL - 2026
JO - Structural Control and Health Monitoring
JF - Structural Control and Health Monitoring
IS - 1
M1 - 5553171
ER -