This paper presents a framework for nonlinear system identification of civil structures using sparsely measured dynamic output response of the structure. Using a sequential maximum likelihood estimation (MLE) approach, the unknown FE model parameters, the measurement noise variances, and the input ground acceleration time histories are estimated jointly. This approach requires the computation of FE response sensitivities with respect to the unknown FE model parameters (i.e., FE parameter sensitivities) as well as the FE response sensitivities with respect to the values of the input ground acceleration at every time step (i.e., FE input sensitivities). The FE parameter and input sensitivities are computed using the direct differentiation method (DDM). The presented output-only nonlinear FE model updating method is validated using the numerically simulated seismic response of a realistic three-dimensional five-story reinforced concrete building structure. The simulated building responses to a horizontal bi-directional seismic excitation is contaminated with artificial measurement noise and used to estimate the unknown FE model parameters characterizing the nonlinear material constitutive laws of the reinforced concrete, as well as the root mean square of the measurement noise at each measurement channel, and the full time history of the seismic base acceleration. The method presented in this paper provides a powerful framework for structural system and damage identification of civil structures, when the input excitations are not measured, are partially measured, or the measured input excitations are erroneous.
|Número de páginas||6|
|Estado||Publicada - 2017|
|Evento||10th International Conference on Structural Dynamics, EURODYN 2017 - Rome, Italia|
Duración: 10 sept. 2017 → 13 sept. 2017
Nota bibliográficaPublisher Copyright:
© 2017 The Authors. Published by Elsevier Ltd.
- Bayesian Inference
- Damage identification
- Input Estimation
- Joint Input
- Model updating
- Nonlinear finite element model
- Structural health monitoring
- System identification