Seismic response evaluation of a five‐story buckling‐restrained braced frame using multi‐element pseudo‐dynamic hybrid simulations

Abstract

Nonlinear time history analysis relies on accurate modeling of the critical structural components and their complex interaction with the structure. Previous research indicates that calibration of numerical models can be affected by several factors, including the loading protocols. It is, therefore, critical to study previously developed and calibrated numerical models under more realistic loading histories, and determine whether the calibration process, loading protocols, and the numerical model themselves are adequate for achieving the desired level of accuracy. High fidelity benchmark system-level experimental-based simulation results could allow for a more holistic assessment of such questions.The University of Toronto Ten Element Hybrid Simulation Platform (UT10) was developed to produce such benchmark test results using hybrid simulations with multiple experimental elements subjected to realistic earthquake loads. This paper presents the first such experiment in the UT10 with multi-element and single-element experimental hybrid simulations on a five-story steel structure with buckling-restrained braces, representative of systems with a stable yielding hysteretic response. An adjustable yielding brace system was developed to capture the response of buckling-restrained braces' yielding core. The implications of modeling choices, such as using commonly available models in BRBFs, are studied. The experimental results are then presented and compared with numerical results. The limitations of existing models are identified. Such experimental results can be used by subsequent studies to improve the calibration of numerical models and allow for the development of more robust models, while also justifying the need for new loading protocols that could be used in the calibration process.

Uncertainty quantification in the calibration of numerical elements in nonlinear seismic analysis

“…However, previous research has shown that minor errors between calibrated models and reference data that are generally accepted by researchers may lead to amplified errors at the system level structural dynamic response. 1,16,17 Zsarnóczay and Baker 1 defined calibration relevance as the capacity of the calibration method to focus on such relevant features of the structural model. They studied calibration relevance by comparing the calibration error at the component level and global error at the system level within the framework of virtual experiments.…”

“…Detailed modeling assumptions can be found in Mojiri 26 . For the reference models, Steel4 parameters were calibrated using component cyclic tests on an AYB 17 as shown in Figure 2. The corresponding Steel4 parameters are listed in Table 2.…”

“…The underlying assumption is that a best‐fit component model is accurate enough and will have the same level of accuracy in predicting the response of a structure when the component is assembled in a structural system subjected to seismic ground motions. However, previous research has shown that minor errors between calibrated models and reference data that are generally accepted by researchers may lead to amplified errors at the system level structural dynamic response 1,16,17 . Zsarnóczay and Baker 1 defined calibration relevance as the capacity of the calibration method to focus on such relevant features of the structural model.…”

Impact of loading rate during hybrid simulation on seismic response of steel structures

“…For example, the time scale t 10 corresponded to a PsD test with a time step of 1 ms in the numerical model and 10 ms ramps to load the physical specimen. In PsD tests, the restoring force of the physical specimen is typically measured at the end of the displacement ramp to minimize errors due to strength relaxation during the hold phase 41,42 . To be consistent with the actual test, the same procedure was followed in this parametric study. t 100 : similar to the case t 10 , the actual time scale was extended by 100 times.…”

Mechanical Modeling and General Analytical Solution for the Dynamic Buckling of Plane Structures Using a Beam-Column Element with Varying End Restraints

“…Previous research on the behaviour of connections shows that such assumptions are not entirely valid and do not represent the real behaviour of nodal connections [3][4][5][6][7][8][9]. They have been adopted in practice for their simplicity in the analysis and design of structures.…”