The Virtual Test Of Asphalt Mixture Stiffness Moduli Based On DEM

Research on asphalt mixture has limited by analysis tools and only empirical methods have been commonly used for a long time. These empirical methods were based on phenomenology, where the macroscale performances of asphalt mixtures were analyzed through laboratory or field tests and making a lot of hypothesises. Obviously, these methods are expensive and time-consuming, and their results are variant and unrepeatable. A direct result from these methods is the seperate beween the design system and road performances of asphalt mixtures. The overall macro- mechanical behaviors of the asphalt mixture are determined by the micromechanics within the cemented particulate system. Based on the heterogeneous multiphase nature of aspahlt material, it appears that a micromechanical model would be best suited to properly simulate such a material. Therefore, this dissertation research presents a micromechanical methodology to predict asphalt mixture stiffness and analyze the effects of its macro-scale properties on the overall mechanical performances of the mixture with discrete element method. Asphalt mastic (fine aggregates, fines, asphalt binder, and air voids) and aggregates are considered as the two basic constituents of asphalt mixture. Random polygon (2D) and polyhedron (3D) algorithms were developed to build the microstructure of asphalt mixture, where the randomly created polygon or polyhedron particles represent aggregates, and the inter-place of aggregate particles is filled with the uniformed balls bonded with viscoelastic contact model to represent asphalt mastics. Laboratory tests were completed including DSR (Dynamic Shear Rheology) tests of asphalt mastic and Uniaxial compressive dynamic modulus tests of asphalt mixtures to serve two purposes: (1) privide input parameters for discrete element simulation; and (2) calibrate simulation results with tests of asphalt mixtures.The contributions of his dissertation work include: (1) an algorithm was developed for randomly creating the irregular polygon or polyhedron particles, with which gradation, shape and distribution of aggregates can be considered. (2) the micromechanical model of asphalt mixture were developed, where the user-written Burger's model and contact stiffness model are employed to represent viscoelastic behaviors of asphalt mastic and the elastic properties of aggregates respectively, and the flip and bonding models are used to consider the strength properties at contact points. (3) Static tri-axial tests of graded stones were simulated and calibrated. The micromechanical properties of graded stones were evaluated by calculating the standard parameter: resilient moduli (MR). (4) Review of the research on dynamic modulus of asphalt mixtures was conducted. The research history, current efforts and commonly used methods were summarized in this dissertation paper. (5) Simulation of asphalt mixtures was conducted, and the dynamic modulus and phase angles were predicted. The process of the dynamic modulus virtual tests can be monitored with the software developed in this research.The performances of asphalt mixtures were investigated with discrete element method by analysis of microstructure in this dissertation work. This research effort cuts a new way for mixture design and damage mechanism analysis of asphalt pavement: not only avoids a lot of laboratory tests, but also gives a more detailed analysis on the micro-scale behaviors under the outer loading conditions.