地震作用下土体液化准则的区域变异性研究

China is one of the countries that is most threatened by the earthquake. Earthquake induced soil liquefaction is one of the common causes that damaged the infrastructure in the seismic zone. Currently, the liquefaction criterion is often developed assuming that soils in different regions follow the same liquefaction criterion. However, recent studies in earthquake engineering show that such an assumption is not valid, and the liquefaction criterion developed based on data in one region may not be applicable to other regions. How to construct the liquefaction assessment criterion considering its inter-region variability is a pressing problem of fundamental importance in liquefaction research. In this project, field tests will be conducted to convert the liquefaction case histories from Mainland China to international format. As such, the case histories can be incorporated into the international database to increase the impact of Chinese database on the liquefaction evaluation of soils. Through multi-level Bayesian analysis, a liquefaction model considering the inter-region variability will then be developed and fitted by the compiled database. How to solve the Bayesian formulation will also be investigated. With the calibrated model, factors affecting the inter-region variability of the liquefaction model will also be assessed, and the characteristics of the inter-region variability will be examined. Finally, the liquefaction assessment criterion will be developed for different regions considering the inter-region variability of the liquefaction models. This research will provide data, theory, and methodology for developing the next generation liquefaction models and liquefaction assessment criteria, and thus substantially enhance our capability of liquefaction assessment.

我国是世界上受地震灾害威胁最严重的国家之一。地震引发的土体液化可引起基础设施的严重破坏。现有理论多假设不同区域土体遵循相同的液化准则。近年研究表明，土体的液化准则具有明显的区域性，基于一个地区液化案例建立的液化准则对其它地区未必适用。如何建立区域性的土体液化准则是地震液化灾害防治中急需解决的基础科学问题。本项目拟通过现场试验和理论分析，对中国大陆液化案例进行完善，将其纳入国际液化案例库，从基础数据层次提高我国案例在液化研究领域的影响力；通过分层贝叶斯分析原理，构建考虑区域变异性的土体液化预测模型，提出相应的求解方法；在此基础上，研究土体液化模型的区域变异性规律及其影响因素，建立考虑区域变异性的土体液化准则。研究将为地震作用下土体液化灾害的预测和评价提供理论依据和技术支持。

我国是世界上受地震灾害威胁最严重的国家之一。现有理论多假设不同区域土体遵循相同的液化准则。近年研究表明，土体的液化准则具有明显的区域性，基于一个地区液化案例建立的液化准则对其它地区未必适用。本项目通过现场试验和理论分析，提出了国标锤击数和美标锤击数的转换关系，建立了全球标准液化案例库；通过模型误差分析，对国标液化准则的误差进行了标定，并提出了相应的改进方法；通过分层贝叶斯分析原理，构建了考虑区域变异性的土体液化预测模型，提出基于杂交马尔可夫链的高效求解方法；以国标液化准则和Youd法液化准则为例，揭示了土体液化模型的区域变异性规律及其影响因素，建立了考虑区域变异性的土体液化准则。研究为地震作用下土体液化灾害的预测和评价提供了理论依据和技术支持。