FDTD Simulation for Moisture Asphalt Pavement Thickness and Density Estimation Utilizing Ground Penetrating Radar

Abstract

Ground penetrating radar (GPR) has the potential to estimate the thickness and density of asphalt pavement during compaction. However, the surface moisture sprayed by the compactor interferes with the accuracy of data collection significantly. This study proposed an approach based on the extended common midpoint (XCMP) method to minimize the effect of surface moisture. Both the numerical simulation of finite-difference time-domain (FDTD) and laboratory experiments were carried out to study the effect of the surface moisture on the GPR signal. Then, three FDTD models with different incident angles of GPR signal were established, and the difference of time intervals obtained from dry and moisture pavements with each model was studied to propose a proper antennas installation mode. Finally, the thickness and density estimated using the proposed method and surface reflection method were compared to validate the accuracy of the proposed approach. The results show that: 1) FDTD models were verified to simulate the interaction of GPR signal with moisture pavement effectively; 2) the time interval of the GPR signal between the surface and bottom of AC layer increased as the thin wet layer dielectric constant grew, and remained unaffected by the electric conductivity of the thin wet layer; 3) the average error of thickness and density predicted utilizing the proposed method were less than 1.3% and 2.4%, respectively, undercomplicated compaction conditions. This study notes that compaction monitoring in real time could benefit from the proposed method.