Bulletin of Earthquake Engineering最新文献

Two new structure-specific scalar intensity measures for plane reinforced concrete moment resisting frames under far-fault ground motions are proposed. These intensity measures, of the spectral acceleration and spectral displacement type, are characterized as multi-modal and multi-level. They encompass the effects of the first four natural periods and are defined for four performance levels, including considerations of inelasticity up to the collapse prevention level. This is achieved with the aid of equivalent linear modal damping ratios previously developed by the authors for performance-based seismic design purposes. These modal damping ratios, dependent on period, soil type, and deformation, are associated with the transformation of the original multi-degree-of-freedom (MDOF) nonlinear structure into an equivalent MDOF linear one. The proposed intensity measures are conceptualized to be simple and elegant, incorporating all the aforementioned features rationally, without the artificial combination of terms, definition of period ranges, or addition of coefficients determined by optimization procedures. This approach sets it apart from existing measures that attempt to account for multiple modes and inelasticity. A comparison of the proposed intensity measures against ten of the most popular existing ones in the literature, focusing on efficiency, practicality, proficiency, scaling robustness and sufficiency, demonstrate their advantages.

Significant damages to precast wall-slab-wall (WSW) systems due to past earthquakes in near-field zones has been reported in the literature. This led to research on the seismic behavior of precast structures. Most of them concentrated on precast framed structures. Comparatively, fewer studies have been conducted on WSW systems, especially in exploring their performance in near-field earthquakes. This study focuses on the analysis of a 5-story precast WSW structure and the corresponding monolithic WSW structures under near-field (NF) and far-field (FF) earthquakes. The normalized backbone curves (M-θ curves) for precast and monolithic wall-slab connections were modeled using link elements at the slab-wall interface. A default plastic hinge is assigned at a distance of 0.1 L from the slab-wall interface. Three types of earthquakes were considered: far-field (FF), near-field forward directivity (NFD), and near-field fling step effect (NFFE). Nonlinear time history analysis (NLTHA) is performed in the computer program SAP2000 using an ensemble of 7 different earthquake records for each type. The earthquake records are normalized for three levels of peak ground acceleration (PGA): 0.4 g, 0.6 g, and 0.8 g. The responses of interest include top story displacement (TSD), maximum inter-story drift ratio (MIDR), base shear (BS), and maximum acceleration (MA). Comparative studies utilized the ensemble average of responses. The findings reveal that the theoretical analysis of precast frames shows greater vulnerability compared to conventional monolithic frames (as commonly practiced without specifying M-θ curves at the slab-wall interface). Moreover, NFFE led to increased top story displacement and MIDR responses in all types of precast and monolithic WSW structures under study.

This study investigates the seismic behaviour of a special mixed reinforced concrete-steel structure that supports an oil refinery reactor. The structure is 64.90 m tall and consists of three parts: (a) a reinforced concrete frame basement; (b) a steel braced frame that supports the oil reactor and (c) the steel reactor itself. A three-dimensional model of the structure is created to perform static non-linear (pushover) analyses in order to obtain the capacity curves and understand the overall inelastic behavior of the structure. The results of the pushover analyses reveal that the structure exhibits similar inelastic behavior in both horizontal directions and satisfies the capacity design principles. The structure exhibits limited ductility considering the fact that has been designed with a behavior factor of q = 1.5 and primary damages are expected mainly in concrete members. Subsequently, dynamic non-linear time-history (NLTH) analyses are performed utilizing the three translational components of three seismic motions recorded during past earthquakes. These results involve: (i) the maximum values for displacements, accelerations and base shears; (ii) the maximum stresses at critical points of the oil refining reactor and (iii) the formation of plastic hinges at columns, beams and braces of the structure. Contrary to pushover analyses, NLTH analyses revealed the development of plastic hinges, hence seismic damage, that do not follow the desirable formation pattern. Moreover, the accelerations and displacements observed are expected to cause failure of the piping and mechanical equipment, while local failure of the high-stress areas of the shell of the reactor may be possible. Localized strengthening might be necessary to avoid repair works and downtime after such seismic event.

The Italian seismic code provides a simplified approach to account for the effect of local seismostratigraphical configuration on the expected ground motion. This approach, common with other seismic codes, provides specific ‘soil factors’ as a function of a set of reference subsoil conditions (soil classes): these factors are considered in 1D subsoil configurations to modify the uniform probability hazard spectrum deduced from probabilistic seismic hazard at reference soil conditions. It is inferred that, to provide a coherent management of uncertainty affecting the response spectrum to be used for the design, the contribution of uncertainty affecting soil factors must be carefully considered to avoid biases in the hazard evaluation. In the present study, variability of soil factors representative of each soil class has been explored by numerical simulation relative to many seismostratigraphical configurations inferred from seismic microzonation studies available in Italy relative to 1689 municipalities. This analysis shows that variability of soil factors is of the same order of magnitude of variability affecting reference response spectra, which implies that the former cannot be neglected as presently happens in the common practice. It is also shown that neglecting this contribution can lead to underestimate the impact of subsoil configuration on the regularized response spectrum provided by the norm, in particular, in the short period range.

In past seismic events, earthquakes have often caused significant damage to buildings. It is noteworthy that most of the existing buildings are reinforced concrete structures. Therefore, in order to mitigate the damage caused by earthquakes, it is important to conduct damage assessment of reinforced concrete structures. Considering that damage at the material level is the fundamental cause of component and structural performance degradation, indices based on material damage often have advantages in reflecting and evaluating component and structural damage. This paper proposes a damage constitutive model for concrete based on existing research results. Then, aiming at the shortcomings of current research on steel bar damage constitutive models, a steel bar damage constitutive model under cyclic loading is proposed, reflecting various failure modes of steel bars under seismic actions. Based on this, a multi-level damage index system from materials to components to structures is established. Through multi-level experimental simulations and finite element analysis, the accuracy of the proposed damage indices is validated, and performance indices for components and structures are provided. These indices can effectively reflect the damaged state of components and entire structures and can be used to guide seismic design, damage assessment, and strengthening design.

In previous earthquakes, a significant number of adjacent buildings in a series have been damaged due to collisions. Pounding between adjacent structures in a series causes them to inflict multiple blows on one another, which is a complex type of collision. Previous studies have produced inconsistent or conflicting research conclusions due to various parameters of buildings and excitation. It is challenging to determine a universal law of collision reactions between adjacent buildings in a row. To address the complexity of these parameters, the dimensional analysis method is used. This work establishes a mathematical model for the dimensionless collision response of adjacent structures in a row. The layout of the structures is considered through three different configurations, and the effects of unilateral and bilateral collisions are compared. The analysis also considers the impacts of the frequency ratio, mass ratio and gap size of the oscillators. According to the impact of pounding, the displacement and velocity responses of the outer structures with low mass and stiffness can be divided into multiple spectral regions based on the frequency ratio of the structure and excitation. The effects of the mass ratio and frequency ratio on the responses of the outer flexible structures are correlated with the spectral regions. The results indicate that placing a structure with a small mass and stiffness outside is dangerous, since it causes a much larger pounding force and displacement of the outer structure. Compared with the unilateral impact response, the bilateral impact response induces a smaller displacement of the middle structure with a slight mass and stiffness.

Under earthquake loads, reinforced concrete chimneys with openings are prone to stress concentration and damage around these openings, possibly leading to structural collapse. In this paper, the stress concentration factor was proposed to quantitatively analyse the range of stress concentrations around openings under various parameters and identify the most significant coefficients affecting the stress concentration range. 9 Groups of numerical analysis models for chimneys with different parameterized openings were established, and more than 200 pushover analyses were conducted. The effects of the chimney wall thickness, wall diameter, opening size, and spacing between openings on the stress around the openings were investigated. Recommendations for limiting opening sizes were given, and a method for designing reinforcement steel bars for strengthening was proposed. The numerical results indicated that the central angle of a single opening cross-section should be less than 70°, and the total central angle of multiple openings should be less than 140°; the spacing between openings should not be less than 0.5 times the width of the opening; the range of the steel reinforcement range should be 4.5 times the wall thickness for circular openings; and for rectangular openings, it should be 3.5 times the wall thickness. The added reinforcement steel should have a reinforcement ratio for strengthening of 1.3 times that of the original reinforcement ratio. Finally, through dynamic analysis, it was verified that the opening reinforcement design method proposed in this paper can effectively reduce the stress concentration around openings.

This special issue focuses on the seismic performance of adjacent interacting masonry structures, particularly in historical European city centers. The AIMS project provided unique data on interacting masonry structures through large-scale shake table tests on two adjacent half-scale stone masonry buildings. The experimental campaign was accomapgnied by a blind prediction study where participants modeled the aggregate’s seismic response. Findings highlight challenges in accurately predicting displacement demands and failure modes, providing critical insights for improving future modeling techniques for masonry buildings.

The existing Reinforced Concrete (RC) buildings stock is often characterized by a significant seismic vulnerability, due to the absence of capacity design principles, even in regions with high seismic hazard, such as Italy. Approximately 67% of existing RC buildings in Italy have been designed without considering seismic actions (GLD), resulting in very low transverse reinforcement amount in beams and, particularly, in columns. Additionally, beam-column joints typically totally lack stirrups. Consequently, shear failures under seismic actions are very likely for this pre-code building typology, often limiting their seismic capacity. However, the assessment of shear failures in beams/columns or joints varies significantly from code to code worldwide. The main goal of this work is to quantify the impact of different code-based brittle capacity models on the seismic capacity assessment and retrofit, focusing on GLD Italian pre-1970 RC buildings. This comparative analysis is carried out by first considering three current codes, emphasizing their, even significant, differences: European (EN 1998-3-1. 2005), Italian (D.M. 2018), and American (ASCE SEI/41 2017) standards. Then, shear capacity models prescribed by the current drafts of the next generation of Eurocodes are implemented and compared to the current models. The assessment includes: (i) a parametric comparison among models; (ii) the evaluation of case-study buildings capacity in their as-built condition and after shear strengthening interventions. The latter is performed on 3D “bare” models, due to the lack of practical guidance in most codes on modelling masonry infills.

Liquefaction can significantly alter the ground response. However, no existing design spectrum accounts for the severity of soil liquefaction. This work aims to develop correction factors that can be used to adjust code-based design spectra to reflect the specific liquefaction susceptibility of a site. The correction factor is derived as the ratio of response spectra calculated by two types of 1D nonlinear site response analyses: effective stress analysis, which can model porewater pressure (PWP) generation, and total stress analysis. We considered seven real profiles and 200 motions in our analysis. Four combinations of soil nonlinear models and PWP generation models are also utilized to account for epistemic uncertainties. Results show that the response spectral ratio for liquefied sites typically falls below one for periods less than 1–2 s and rises above one for longer periods. Meanwhile, the response spectral ratio reflects the overall liquefaction susceptibility influenced by PWP, factor of safety, and liquefiable layer depth, while the liquefaction potential index (LPI) captures their complex interplay. Accordingly, we propose four LPI-dependent factors: three correction factors for peak ground acceleration, 0.2 s spectral acceleration (Sa), and 1.0 s Sa, and a long-period adjustment factor applicable for periods exceeding 1 s. The correction factors linearly decrease with increasing LPI, while the adjustment factor exhibits the opposite trend. A design spectrum for a liquefiable site can be readily constructed by adjusting the code-based design spectrum using the proposed correction factor, as illustrated in the example. This approach is applicable as long as LPI is available from a simplified liquefaction analysis or a liquefaction hazard map.