基于数字图像的大空隙沥青混凝土离散元模型  

Discrete element model for porous asphalt concrete based on digital images

作　　者：马翔[1] 袁则瑜 陈满军[2] 周培圣 MA Xiang;YUAN Zeyu;CHEN Manjun;ZHOU Peisheng(College of Civil Engineering,Nanjing Forestry University,Nanjing,Jiangsu 210037,China;Kunshan Construct Engineering Quality Testing Center,Kunshan,Jiangsu 215337,China)

机构地区：[1]南京林业大学土木工程学院,江苏南京210037 [2]昆山市建设工程质量检测中心,江苏昆山215337

出　　处：《江苏大学学报（自然科学版）》2019年第6期734-739,共6页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金资助项目(60472120);住房城乡建设部科学技术项目(2015-K4-017);南京林业大学高层次人才科研启动基金资助项目(G2014014);江苏高校优势学科建设工程项目

摘　　要：为了从细观角度深入分析大空隙沥青混凝土性能,利用CT扫描获取PAC-13大空隙沥青混凝土内部图像,将三维重构的切片图像和CT横向扫描图进行阈值分割,通过体视学理论选取与真实级配最相近的图像作为集料、空隙结构提取对象,建立了像素边长为0.5 mm的离散元模型,并进行单轴压缩虚拟测试.结果表明:三维重构切片图的级配误差比横向扫描图级配误差小了约20%,重构的切片图级配更具代表性;该模型实现的虚拟单轴压缩试验强度值与室内单轴压缩试验强度值相差小于3%,该模型具有很好的可靠性,这种数字图像建模的方法为研究大空隙沥青混凝土内部细观结构提供了参考.To analyze the characteristics of porous asphalt pavement in micro-perspective way,X-ray CT technology was used to obtain the internal images of PAC-13, and the image closest to the real gradation was selected to extract aggregate and void structure by stereology after image segmentation. The discrete element model(DEM) was established with pixel side length of 0.5 mm, and the uniaxial compression virtual experiment was carried out. The results show that the gradation error of 3 D reconstructed slice image is nearly 20% smaller than that of transverse scan image, and the reconstructed slice image gradation is more representative. The compressive strength of virtual uniaxial compression test performed by the model is less than 3% different from that of real uniaxial compression test. The model has good reliability,which can provide research basis for studying the internal mesoscopic structure of porous asphalt concrete.

关 键 词：大空隙沥青混合料 阈值分割 离散元 X-RAY CT扫描 单轴压缩 

分 类 号：U414[交通运输工程—道路与铁道工程]