【作者】 钟菊芳；

【导师】 吴胜兴； 胡晓；

【作者基本信息】 河海大学 ， 结构工程， 2006， 博士

【摘要】 合理的地震动输入是保证结构抗震设计结果可靠与经济的必要条件。目前国内外普遍采用的反应谱法没有考虑地震动持时及空间变化性的影响；拟合目标反应谱的人造地震动方法不能得到频率非平稳加速度；国内规范在加速度峰值的选取方面存在不相衔接的现象；一致概率反应谱并不是同一地震事件的真实反应谱，该谱的中、长周期值不合理显著偏大，造成抗震设计十分困难。本文针对本文针对地震加速度峰值特性、时频谱的衰减特性及地震动参数的空间变化性等方面进行了深入研究，并在此基础上探讨了综合考虑地震动三要素影响的设定地震方法和工程场地非平稳加速度时程的合成方法；最后将研究结果应用于新疆克孜尔坝址地震动输入参数的确定。主要内容及成果如下： (1)通过对地震加速度记录放大系数谱的计算分析，给出了基岩和土层记录的水平和竖向地震分量的有效峰值加速度(EPA)的计算式。通过对峰值加速度(PGA)和EPA统计分析表明，PGA与EPA间存在强线性和对数线性相关性，可根据相关性方程进行PGA与EPA间的相互转换或预测，便于国内规范的配套使用。通过影响因素分析发现，衰减关系式的选取、年超越概率的大小、场点与潜在震源的位置关系及潜在震源的震级上限的大小等都将直接影响PGA和EPA取值及两者的比例关系。 (2)根据中国大陆与美国西部在地震构造、地震成因等方面的相似性，利用美国西部基岩加速度强震记录资料，建立了适用于中国大陆的时变功率谱和能量等效速度谱的衰减关系式。分析了震级、震中距对时变功率谱及能量等效速度谱的影响规律。 (3)应用通过分析局部空间位置变化对地震动参数影响显著程度来研究地震动参数空间分布规律的方法，建立了地震动参数随局部空间位置坐标变化的随机预测模型。结果表明：空间位置坐标的变化对不同的地震动参数或模型参数的影响规律及影响程度不相同，不同的地震动参数具有不同的空间分布形式。 (4)提出了由EPA确定最大贡献潜源、由0.2s周期点的能量等效速度谱确定设定地震的震级和震中距、并根据地震构造条件确定设定地震的具体位置的设定地震方法。该方法综合考虑了地震动三要素—幅值、频谱和持时对设定地震结果的影响。 (5)探讨了根据时变功率谱衰减规律、地震动参数空间变化模型及设定地震结果进行多点输入非平稳加速度时程合成的方法。 (6)利用本文提出的方法，对新疆克孜尔坝址所在的地震研究区进行了地震危险性分析和设定地震研究，确定了坝址4种超越概率下的地震动输入参数——加速度峰值、加速度反应谱及加速度时程。并对依据不同参数进行设定地震所取得的结果进行了比较分析，验证了本文设定地震方法的合理性。更多还原

【Abstract】 The reasonable earthquake ground motion input parameters are necessary to insure the correct designed results. However, nowadays, the response spectrum method does not take the influence of the temporal and spatial variability of the ground motion on the structure seismic response into account, and the ground motion frequency nonstationarity is not considered properly. Besides, there is in concordance in selecting acceleration peak in different domestic specifications. In this paper, some problems on the characteristics of ground motion acceleration peak, spatial variabilities and attenuation laws of the evolutionary power spectrum and the absolute elastic input energy equivalent velocity spectrum have been studied. Then, the methods of generating the scenario earthquake and synthesizing nonstationary strong ground motion have been discussed. Finally, the method was applied to determine the earthquake input parameters of Kezier dam site in Xinjiang. The main results of this paper are as following:(1) The calculated methods of effective peak ground acceleration (EPA) were put forward by analyzing the acceleration average magnified coefficient spectrum. The statistical results indicated that there was strong linear and logarithm linear correlation between peak ground acceleration (PGA) and EPA. The feasibility and rationality about using EPA to replace PGA were also demonstrated. The main effect factors on PGA, EPA and the proportion of PGA and EPA were discussed. The results suggested that the attenuation relationship, annual exceeding probability, the locations of site and potential earthquake sources and the maximum magnitude of potential earthquake source had directly influence on the value of PGA, EPA and PGA/EPA.(2) Based on the rock strong motion dataset in western North America, the attenuation relationships of evolutionary power spectrum and energy equivalent velocity spectrum were developed according to the similarity in geology, seismology and regional tectonics et al between China and western North America. The results revealed that models presented in this paper fitted the record’s values better than those in correlative literature.(3) The method being used to analyze ground motion spatial variabilities by analyzing the influence of the variation of spatial distances on ground parameters was studied, and the stochastic prediction models were suggested as well. The results proposed that different parameters had different spatial distribution forms.(4) A scenario earthquake method which integrated the amplitude, duration and frequency of ground motion was presented. In this method, the dominant potential earthquake sources, which had the largest contribution to the site, were chosen from earthquake sources according to the EPA attenuation relationship. The magnitude and hypocentral distance of scenario earthquake were determined by energy equivalent velocity spectra attenuation relationship at 0.2 seconds periods within the dominant potential earthquake sources, and the specific更多还原