Earthquake Engineering & Structural Dynamics最新文献

A significant challenge in performance-based seismic design (PBSD) is to manage uncertainties in a cost-efficient manner. This highlights the necessity of integrating PBSD with a robust design optimisation (RDO) methodology. The objective is to find optimal solutions that are not only safe and cost-effective but also minimally sensitive to variations in input noise parameters such as record-to-record variability (RTRV) of ground motions and uncertainties related to structural properties. In this study, for the first time, an RDO formulation is tailored to the PBSD of reinforced concrete (RC) bridge piers. The developed optimisation framework is applied to the PBSD of RC piers following the specifications of fib Model Code 2010. The obtained results provide valuable insights into the trade-offs between economic costs and robustness of RC pier design solutions. They also underscore the importance of the proposed RDO method in decision-making processes, highlighting solutions that may be slightly more expensive but significantly more robust compared to the most economical options.

The dynamic behavior of seismically isolated structures is governed by the force-deformation response of the isolation devices. Consequently, significant efforts have been made to accurately simulate the behavior of different types of devices. High damping rubber bearings (HDRBs) are among the most widely manufactured and used isolators in practice. Given the internal structure of these devices and the characteristic behavior of the rubber compound, HDRBs show highly nonlinear behavior with strong coupling between deformation directions, which is challenging to simulate numerically. Capturing these complex multiaxial interactions is essential for reliably predicting device behavior and ensuring the dynamic stability of the isolation system during seismic events, therefore, a holistic multiaxial modeling approach is critical. This study presents a robust and sufficiently accurate numerical model for simulating the multiaxial behavior of HDRBs under large deformations. This elaborate model includes: bidirectional shear response that accounts for stiffness degradation with load-direction dependency, including scragging (long-term degradation) and Mullins effect (short-term degradation), and temporary hardening; coupling between axial and shear response, including axial stiffness softening due to lateral displacement and shear stiffness variability due to axial load variation; axial instability due to large compressive loads; and cavitation under tensile forces. The proposed model is validated using a wide range of load patterns applied to an HDRB, as well as experimental results from the literature. The proposed model demonstrates good agreement with experimental data, accurately simulating HDRB responses across diverse validation tests, including double bidirectional shear tests in rotated directions, cyclic shear response under different axial loads, tensile loads, bidirectional deformation history with an elliptical orbit, extremely large deformations (beyond the design limits), and dynamic analyses. The results show that the model provides reliable predictions of the static and dynamic behavior of HDRBs under different load patterns, including deformations until the onset of failure. The proposed model has been implemented in OpenSees and is openly available at the supplementary repository https://github.com/JAGallardo1992/HDRB_model.

Rocking of rigid blocks is a motion pattern of many nonstructural components. This paper investigates the rocking control of rigid blocks with rotary tuned mass damper (RTMD). A 2-DOF analysis model is proposed for the RTMD-controlled rocking system, and the rocking motion equations are established by the Lagrangian method. The free rocking responses of the RTMD-controlled system are analytically solved, which is further verified in comparison with the numerical simulation. The RTMD is capable of involving positive modal stiffness in the rocking system due to the torsional stiffness. The tuning stiffness and damping of RTMD are key parameters for vibration control. The roles of RTMD tuning parameters and inertia ratio are investigated through analyses of free rocking responses and overturning accelerations. In conclusion, the design period of RTMD corresponding to the fastest decay of free rocking motion and the maximum increase in overturning acceleration is essentially the same, and the RTMD design period can be quickly obtained through analytical solutions of free rocking response. Undamped RTMDs can significantly improve the maximum overturning acceleration, while increasing damping further enhances this improvement, more notably under high-frequency excitation. In addition, RTMDs contribute mainly to reducing the occurrence of overturning with impact.

The economic, functional, and other losses caused by nonstructural components play a crucial role in earthquake risk assessment of buildings. Although the impact of nonstructural components on seismic damage assessment has been recognized, the correlation of damage among nonstructural components has not been adequately taken into account in current research. Since fragility analysis of nonstructural components is an important part of earthquake loss evaluation, this study proposes a modified fragility method that reflects the damage correlation of nonstructural components. Due to the complexity of the nonstructural systems, most current research focuses on individual components. To address this gap, a calculation method for determining damage correlation using the existing fragility database is introduced. Using three prototype RC moment-resisting frames of 4-, 8-, and 12-story heights, this study investigates the impact of damage correlation between structural and nonstructural components on loss assessment. The results highlight that damage correlation significantly influences the failure probability of nonstructural components under certain conditions, underscoring the necessity of considering such correlations in loss assessment. The proposed method is simple, effective, and provides a valuable tool for improving the accuracy of earthquake loss evaluations.

Previous seismic fault inversions have indicated that the multi-fault rupture occurring at dense fault regions is one of the leading causes in producing strong earthquakes. The multi-fault rupture focal mechanism directly affects the characteristics of ground motions and causes more severe damage to bridges, considering the cumulative damage. The seismic responses of bridges under sequential earthquakes, that is, mainshock-aftershock sequences, have been extensively studied. However, the effects of multiple sequential ground motions induced by multi-fault rupture in a single earthquake remain unaddressed. This study aims to address this research gap by focusing on how multi-fault ruptures affect bridge responses. A numerical model with two vertical strike-slip faults was developed to simulate the multi-fault ruptures using the spectral element method (SEM). Parametric analyses were conducted to investigate the effects of four main influential factors on the characteristics of the multiple sequential ground motions. The seismic response sensitivity of a multi-span continuous highway bridge under such ground motions was examined. The results demonstrated that multi-fault ruptures in an earthquake can generate multiple sequential ground motions in fault-overlapping regions. Such ground motions can significantly amplify the seismic responses of the bridge, increasing damage by 25% compared to the traditional ground motions with a single sequence. The fault-to-fault distance and angle significantly influence the latter phase of the multiple sequential ground motions. A small fault-to-fault distance and angle have the potential to cause significantly greater seismic damage to the bridge. The location of the nucleation zone and the initial shear stress on the fault plane mainly affect the characteristics of the front phase of the multiple sequential ground motions.

Freestanding building contents represent a large portion of non-structural components in the built environment and therefore receive significant attention. However, research using realistic friction coefficient to investigate their seismic fragility is still scarce. In addition, limited efforts have been made to understand whether a well-established technique (e.g., seismic isolation) is beneficial towards protecting unanchored building objects under floor motions. To address this knowledge gap, this study first presents an analytical formulation for a base-isolated freestanding rigid block system, including: the exact nonlinear equations-of-motion, transitional criteria between different motion modes, as well as the approach to handle impact. The developed analytical model is then validated by comparing responses predicted from a different numerical model in which the penalty stiffness method is used to handle contact and impact. Subsequently, fragility curves of freestanding building contents located at different floors of four code-compliant steel moment-resisting frames are developed under a suite of fourteen pairs of ground motions. These fragility curves consider different slenderness angles of blocks, friction coefficients, floor locations, strength reduction factors, as well as whether or not using base-isolation. The influences of these parameters on the overturning probabilities of freestanding building contents are compared and analyzed.

Pile-supported wharves (PSW) are vital for port infrastructure, making seismic resilience essential for regional economic stability. Shake table testing is fundamental to evaluating PSW seismic performance, with large-scale specimens required to accurately replicate prototype pile materials and pile-deck connections. However, the use of heavy-load soil boxes consumes much of the table's load capacity, limiting high-level PGA seismic excitation. Additionally, the scaling effects of the soil complicate the accurate simulation of pile-soil interactions. To address this, an offline real-time substructure shake table testing (offline RSST) is proposed, eliminating the need for the soil box and enabling high-level PGA excitation for large-scale specimens. In this method, the PSW is divided at the soil surface; the foundation is modeled as a simplified amplifier providing lateral stiffness, and together with the superstructure forms the wharf substructure (WS). The input to the WS is adjusted via an offline FFT/IFFT procedure to replicate the prototype's response. Numerical validation for a batter-pile PSW showed frequency domain errors of 3.1% in deck acceleration and 1.4% in batter-pile axial force. Subsequent 1:10 large-scale shake table tests further validated the method, demonstrating accurate replication of key responses under multi-level excitation and providing a practical framework for seismic performance assessment of PSW.

Ground motion is often recommended to be applied with all possible orientations relative to the plan layout of a structure, and the resulting maximum response for an engineering demand parameter (EDP) over all possible directions of arrival (DOAs) is considered in seismic design. The associated DOA is denoted as the critical orientation, which is expected to differ from one EDP to another. A description of the six-component acceleration time series completely defines the required ground motion inputs for most structures. This paper proposes a framework for the maximum response of an EDP over all possible DOAs and the associated critical orientation using three sets of response history analysis of the structure, followed by nominal post-processing. The proposed formulation is “mathematically exact” for linear-elastic systems. A 5- and 30-story reinforced-concrete (RC) building constituted from moment-resisting frames (MRFs) and recorded six-component ground excitations is considered for illustration. Several EDPs are included in this illustration comparing the associated critical orientations, and the resulting variation is significant. The response of an EDP does not significantly change in the vicinity of critical orientation, which, however, is not true at any other arbitrary orientation. Finally, a couple of RC-MRF buildings are considered to understand the variation in critical orientation if the structure is expected to respond in the nonlinear regime. Interestingly, the critical orientation does not alter significantly owing to this inelastic excursion. Based on this limited investigation, the proposed framework may conveniently be incorporated into routine seismic design to account for the maximum direction shaking of multicomponent seismic excitation.