ABSTRACT

To reveal the fracture damage of asphalt mixture, an image-based numerical approach is proposed in this study considering the distribution of aggregate and asphalt mortar. Both digital image processing technique and discrete element method are applied to simulate the inhomogeneity of specimen and its fracture evolution at low temperature. At the same time, the acoustic emission (AE) activities are measured in laboratory to be compared with the micromechanical responses of numerical models. Results indicate that the developed method is suitable for modelling the numerical specimen of asphalt mixture, taking into account the size gradation and morphology of aggregates in the two-dimensional image. The numerical results of asphalt mixture have a good agreement with that of laboratory tests in cracking distribution and statistical probability. Cracks first appear on the aggregate–mortar interface, located in the upper and middle parts of the specimen, and then the brittle shear failure gradually propagates along the surface. Cumulative probabilities of both AE ringing counts and microfracture number follow the Weibull distribution. The numerical model has a significant advantage in the fracture evolution of asphalt mixture because the eccentric loading is unavoidable in laboratory tests due to personal and mechanical errors.

摘要

为了揭示沥青混合料的断裂损伤，本研究提出了一种考虑骨料和沥青砂浆分布的基于图像的数值方法。采用数字图像处理技术和离散元方法模拟了试样在低温下的不均匀性及其断裂演化。同时，在实验室测量声发射（AE）活动，与数值模型的微机械响应进行比较。结果表明，所开发的方法适用于模拟沥青混合料的数值试样，同时考虑了二维图像中集料的尺寸级配和形态。沥青混合料的数值结果与室内试验结果在开裂分布和统计概率方面有很好的一致性。裂纹首先出现在骨料-砂浆界面，位于试件的上中部，然后脆性剪切破坏沿表面逐渐扩展。AE 振铃计数和微裂缝数的累积概率服从 Weibull 分布。数值模型在沥青混合料的断裂演化中具有显着优势，因为在实验室试验中由于人为和机械误差而产生的偏心载荷是不可避免的。