Determination of complex modulus gradients of flexible pavements using falling weight deflectometer and artificial intelligence

Abstract

The modulus gradient of asphalt concrete (AC) layers is an important feature of flexible pavements. The variation of the modulus with depth results from the synthetical effect of material properties, the service time of pavements, loading and environmental conditions. Since the modulus gradient directly affects critical responses and performance of pavements, the determination of the modulus gradient of AC layers is necessary for the evaluation, maintenance and rehabilitation of flexible pavements. This paper aims to propose a method to obtain layer moduli of flexible pavements at different loading frequencies, which include a power function describing the modulus gradient of AC layers. The method utilizes results from a typical nondestructive test in the field applying the falling weight deflectometer and techniques of the fast Fourier transform, finite element model updating, kriging model and artificial intelligence. The method is validated by comparing layer moduli obtained from the proposed method and other backcalculation softwares.