持续时间(音乐)
结构工程
力矩(物理)
还原(数学)
双线性插值
地震动
强度折减
增量动力分析
非线性系统
航程(航空)
降级(电信)
灵敏度(控制系统)
岩土工程
工程类
数学
统计
物理
有限元法
航空航天工程
电子工程
声学
电信
经典力学
量子力学
几何学
作者
M. A. Bravo-Haro,A.Y. Elghazouli
标识
DOI:10.1016/j.soildyn.2018.08.027
摘要
The influence of ground motion duration on the seismic response of steel moment frames is examined is this paper, with due consideration for cyclic degradation effects. A set of 77 spectrally equivalent pairs of short and long records is utilised in detailed nonlinear dynamic assessments in order to isolate the effects of ground motion duration. The influence of duration is firstly evaluated considering degrading and non-degrading idealised bilinear SDOF systems, for various levels of lateral strength representing practical ranges encountered in design. Subsequently, a sensitivity assessment focusing on the main parameters affecting the response of hysteretic degrading models is carried out through comparative incremental dynamic analysis. Whilst the effect of duration becomes more pronounced with the increase in lateral strength demands, particularly when approaching collapse, the cyclic degradation rate is shown to play a significant role even at lower levels typically associated with design. The performance of EC8-compliant frames indicates a higher probability of collapse when long-duration ground motion records are used, with a typical reduction of about 20% in the collapse capacity, in comparison with short-duration cases. The influence of duration is also examined through collapse capacity spectra, based on the seismic performance of 50 steel moment frames, which show that considerable reduction in the structural collapse capacity of structural systems occurs when relatively long duration records are adopted, for a wide range of dynamic characteristics. This becomes particularly evident in the case of buildings with relatively significant cyclic deterioration rates, where collapse capacity reductions up to 40% due to the influence of earthquake duration are obtained.
科研通智能强力驱动
Strongly Powered by AbleSci AI