Estimation of Resilient Modulus of Unbound Aggregates Using Performance-Related Base Course Properties
Estimation of Resilient Modulus of Unbound Aggregates Using Performance-Related Base Course Properties
复制标题
DOI：
10.1061/(asce)mt.1943-5533.0001147
复制
发表时间：
2015-06-01
影响因子：
3.2
通讯作者：
Lytton, Robert L.
中科院分区：
文献类型：
--
作者：
Gu, Fan;Sahin, Hakan;Lytton, Robert L.
本研究旨在发展一种准确、有效的方法来估算非结合集料的回弹模量。首先，提出了一个新的弹性模量模型，将水分的弹性模量的依赖性，除了在现有的模型中的应力依赖。其次，建立了预测模型，方便准确地确定模型中的系数。为了表征非结合集料的水分依赖性，在模型中加入了饱和度和基质吸力参数。土-水特征曲线（SWCC）用于确定在任何给定含水量下的基质吸力值。该模型的水分依赖性进行了验证，为选定的材料具有不同的水分含量。为了建立该模型系数的预测模型，对20种不同基层材料进行了室内试验和多元回归分析。室内试验包括改进的重复荷载三轴试验和与性能相关的基层性能试验。改进的重复荷载三轴试验方案比现行的试验方案更能适应实际交通荷载作用下基层的受力状态。一系列可重复的和性能相关的基层性质的测量和使用开发的预测模型的基础上，多元回归分析。这些新提出的性能包括亚甲蓝值（MBV），百分细粒含量（PFC），粒度的级配，形状，棱角性和纹理的骨料。所开发的预测模型具有较高的R平方值比那些使用其他基层属性。
This study aims at developing an accurate and efficient methodology to estimate the resilient modulus of unbound aggregates. First, a new resilient modulus model is proposed to incorporate the moisture dependence of the resilient modulus in addition to the stress dependence in existing models. Second, prediction models are developed to conveniently and accurately determine the coefficients in the proposed model. In order to characterize the moisture dependence of unbound aggregates, the degree of saturation and the matric suction parameter are added into the proposed model. The soil-water characteristic curve (SWCC) is used to determine the matric suction value at any given moisture content. The moisture dependence of the model is validated for selected materials with different moisture contents. In order to develop prediction models for the coefficients in the proposed model, laboratory experiments and multiple regression analysis are conducted on 20 different base course materials. The laboratory experiments include the improved repeated load triaxial test and tests to measure performance-related base course properties. A new test protocol is developed for the improved repeated load triaxial test, which is better adapted to the stress state of the base course under the actual traffic load than the current test protocols. A series of repeatable and performance-related base course properties are measured and used to develop the prediction models based on multiple regression analysis. These newly proposed properties include methylene blue value (MBV), percent fines content (pfc), gradation of particle sizes, and shape, angularity, and texture of aggregates. The developed prediction models have higher R-squared values than those using other base course properties.