| The asphalt pavement is regarded as one of the important pattern of high- grade highway. Its construction quality control is becoming the highlight problem during the construction of asphalt pavement. For now, the pattern of quality control on asphalt pavement construction represents as the macro of data result and the single of evaluation index. In most conditions, the result of compaction degree on core specimen is an average value, which cannot reflect the variation on vertical structure. At the same time, the single index of density cannot represent the difference in gradations, air voids and the content of binders. Because the volumetric indicators just represent the mechanical properties of asphalt mixtures indirectly. The methods for analyzing virtual mechanics, such as Finite Element Method and Discrete Element Method, are immature.In this paper, the starting point is the microstructure of asphalt pavement, and the purpose of the study is to put forward the homogeneity analysis method of core specimens’ performance. It mainly utilizes X-ray Computed Tomography technique to acquire the information of internal structure. On one hand, the volumetric information, such as the coarse aggregate and air voids, is evaluated. In this way, the distribution of coarse aggregates and air voids are obtained, and the two indicators, coarse aggregate ratio in section and non-uniformity coefficient, are developed. On the other hand, the evaluation of virtual mechanics performance in asphalt mixtures with finite element method is adopted, and the evaluating indicators for virtual mechanics such as average of the compressive stress and the non-uniformity coefficient of stress are put forward. In short, these indicators could evaluate the homogeneity of asphalt mixture in volumetric and virtual mechanics aspects.In order to characterize coarse aggregate in computed tomography(CT) images with different situation precisely, this paper develop a modified method for the computed tomography(C T) images with dark in the middle and bright in the edge, the OTSU method with circular blocking is adopted; and for the computed tomography(C T) images with even brightness, the AVIZO software is used.To evaluate the volumetric information in asphalt mixture CT images precisely, the Kolmogorov-Smirnov single sample testing method is used to determine the sector scanning degree with 15 degrees, and a statistic method is adopted to determine the least number of slices with 50 pictures per specimen. Based on voxels as the basic element, the three dimension reconstruction of virtual specimens are achieved, and then the virtual compressive test is conducted.Among the different core specimens, the values of volumetric and virtual mechanics indicators are compared. The box-plot is introduced. Compared with the length of box, the value of median, and the number of anomalies, the performance of each indicators is showed. At last, the homogeneity of asphalt mixture in volumetric and virtual mechanics aspects is evaluated. |