Study on the micromechanical response during permanent deformation of asphalt mixtures by discrete element modeling with real aggregate morphology

• Published in: Construction & building materials Vol. 392; p. 131778
• Main Authors: Liu, Yanping, Xie, Jianguang, Zhang, Ye, Wei, Dingbing, Li, Kuan, Song, Jing, Dai, Zexinyu
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.08.2023
• Subjects: Aggregate morphology; Discrete element modeling; Image-processed algorithms; Permanent deformation; Virtual rutting test model; Aggregate morphology; Discrete element modeling; Image-processed algorithms; Virtual rutting test model; Permanent deformation
• ISSN: 0950-0618; 1879-0526
• DOI: 10.1016/j.conbuildmat.2023.131778

•Python image processing algorithms for separating aggregates and asphalt mastic.•2D Python-developed virtual model was proposed to analyze rutting mechanism.•Pioneered the interaction between PFC2D and Python. This study developed a novel 2D virtual rutting model to analyze the permanent deformation mechanism and the micromechanical response of asphalt mixtures. Digital samples of asphalt mixture were image-processed by Python-based algorithms to effectively segment coarse aggregate and asphalt mastic. The Python processed images were imported into PFC2D to establish a 2D virtual rutting model that incorporated the aggregate morphology, the aggregate gradation, and the aggregate void distribution. The feasibility and accuracy of the 2D virtual rutting model were validated by macroscopic laboratory rutting tests. The 2D virtual rutting test had 6.22% higher dynamic stability and 2.12% lower rutting deformation than the laboratory indoor rutting test. As rutting depth increased, the number of effective contacts declined to a steady value. The average Burger contact force between particles within the rutting model increased as loaded. Results demonstrated that the 2D virtual rutting test model was capable of simulating the time-dependent rutting deformation by incorporating Python-coupled algorithms to process aggregates' topography characteristics.