Abstract:The behavior of reinforced concrete (RC) structures under severe demands, as strong ground motions, is highly complex; this is mainly due to joint operation of concrete and steel, with several coupled failure modes. Furthermore, given the increasing awareness and concern for the important seismic worldwide risk, new developments have arisen in earthquake engineering. Nonetheless, simplified numerical models are widely used (given their moderate computational cost), and many developments rely mainly on them. The authors have started a long-term research whose final objective is to provide, by using advanced numerical models, solid basis for these developments. Those models are based on continuum mechanics, and consider Plastic Damage Model to simulate concrete behavior. Within this context, this paper presents a new methodology to calculate damage variables evolution; the proposed approach is based in the Lubliner/Lee/Fenves formulation and provides closed-form expressions of the compressive and tensile damage variables in terms of the corresponding strains. This methodology does not require calibration with experimental results and incorporates a strategy to avoid mesh-sensitivity. A particular algorithm, suitable for implementation in Abaqus, is described. Mesh-insensitivity is validated in a simple tension example. Accuracy and reliability are verified by simulating a cyclic experiment on a plain concrete specimen. Two laboratory experiments consisting in pushing until failure two 2-D RC frames are simulated with the proposed approach to investigate its ability to reproduce actual monotonic behavior of RC structures; the obtained results are also compared with the aforementioned simplified models that are commonly employed in earthquake engineering.

SUMMARYThis paper proposes energy input spectra applicable to seismic design of structures located in lowto-moderate-seismicity regions. These spectra represent the load e ect, in terms of input energy, of the most severe earthquake that the construction might encounter during its lifetime. The spectra have been derived through dynamic response analyses of over 100 ground motion records obtained from 48 earthquakes that have occurred in Spain. An empirical equation for estimating the energy input contributable to damage from the total input energy is also suggested. This equation takes into account both the damping and the degree of plastiÿcation of the structure. Finally, the proposed design energy input spectra are compared with the provisions of the current Spanish Seismic Code and with the response spectra of recent earthquakes that have occurred in Turkey and Taiwan.

No abstract

SUMMARY:This work proposes design energy spectra in terms of an equivalent velocity, intended for regions with design peak acceleration 0.3 g or higher. These spectra were derived through linear and nonlinear dynamic analyses on a number of selected Turkish strong ground motion records. In the long and mid period ranges the analyses are linear, given the relative insensitivity of the spectra to structural parameters other than the fundamental period; conversely, in the short period range, the spectra are more sensitive to the structural parameters and, hence, nonlinear analyses are required. The selected records are classified in eight groups with respect to soil type (stiff or soft soil), the severity of the earthquake in terms of surface magnitude M s (M s  5.5 and M s > 5.5) and the relevance of the near-source effects (impulsive or vibratory). For each of these groups, median and characteristic spectra are proposed; such levels would respectively correspond to 50% and 95% percentiles. These spectra have an initial linear growing branch in the short period range, a horizontal branch in the mid period range and a descending branch in the long period range. Empirical criteria for estimating the hysteretic energy from the input energy are suggested. The proposed design spectra are compared with those obtained from other studies.

No abstract