PI - Y. Richard Kim, Roy H. Borden, and Yasuyuki Horie North Carolina State University "A Unified Approach to Predicting Long Term Performance of Asphalt-Aggregate Mixtures"
The principal objective of the proposed research is to develop test methods and models for predicting long term performance of asphalt concrete that can account for viscoelasticity, damage, volumetric/deviatoric coupling, temperature, and aging. The objective will be accomplished by applying the extended elastic-viscoelastic correspondence principle and the continuum damage theory for describing damage evolution. The experimental results from axisymmetric triaxial tests will be used in determining the multiaxial viscoelastic material functions and finding an explicit form of the constitutive model that will serve as the basis of the performance prediction models. In support of the continuum damage modeling, discrete element modeling will be performed using DM2 code with an objective of establishing a methodology for modeling asphalt concrete behavior on the basis of composition, microstructure, and interfacial properties at the particle level.
The constitutive model will be incorporated into the ABAQUS finite element code, and used to predict the response and performance of laboratory pavements loaded by the Model Mobile Load Simulator. Nondestructive surface wave tests along with dispersion analysis and inversion using Artificial Neural Networks will be performed to determine the change in the modulus of asphalt concrete due to fatigue damage growth. Physical measurements of the change in the asphalt concrete layer thickness will be used to measure permanent deformation growth. The measurements will be used to verify the constitutive model and the finite element analysis. Finally, the mixture and performance data obtained from WesTrack pavement sections will be used to validate the performance prediction models.
