Abstract：
A comprehensive experimental database containing results of a series of dry vacuum-consolidated triaxial compression tests was populated. The tests were performed on sand specimens and conducted under similar experimental conditions, in which specimens\' boundary deformation was captured using a three-dimensional digital image correlation analysis (3D-DIC). The use of a standard triaxial device along with the 3D-DIC technology allowed the specimens\' global and local boundary displacement fields to be computed from start to end of the compression phase. By repeating each test under the same experimental conditions and building the specimens using the same type of sand, the boundary deformation patterns could be identified, and the statistics associated with both global and local displacement fields could be assessed. Making this experimental database available to others should serve to calibrate as well as develop new forward models to account for effects associated with the specimens\' local displacements and material heterogeneity and include statistics to represent a specimen\'s random response. Moreover, this work will serve as a basis for the statistical characterization of spatio-temporal boundary localization effects used to develop stochastic models and machine-learning models, and simulate virtual triaxial tests.
摘要：
建立了一个综合的实验数据库，其中包含一系列干式真空固结三轴压缩测试的结果。试验是在砂样上进行的,试验条件相似,其中使用三维数字图像相关分析(3D-DIC)捕获标本的边界变形。使用标准三轴设备以及3D-DIC技术允许从压缩阶段的开始到结束计算样本的全局和局部边界位移场。通过在相同的实验条件下重复每个测试，并使用相同类型的沙子构建样本，可以识别边界变形模式，可以评估与全球和地方流离失所领域相关的统计数据。将此实验数据库提供给其他人应用于校准和开发新的正向模型，以考虑与标本的局部位移和材料异质性相关的影响，并包括统计数据来表示标本的随机响应。此外，这项工作将作为用于开发随机模型和机器学习模型的时空边界定位效应的统计表征的基础，并模拟虚拟三轴试验。
