Development and validation of a degree of saturation prediction model using time domain reflectometry for compaction control

Abstract

Compaction quality is directly related with the shear strength of soils. Inadequately compacted soils have lower shear strength that decrease the load bearing capacity and increase the differential settlement of the pavement, which may endanger safety and durability of the highways. Although intelligent compaction technology has enhanced the quality control of the compaction, conventional evaluation methods adopting sampling from random locations to determine relative compaction are extensively used in most of the earthwork projects. However, relative compaction may lead misjudgment, since the compaction energy level and the soil type in-place cannot match exactly with the given specifications of an earthwork project. Recent studies have shown that soil degree of saturation (Sr) can be utilized as an alternative control parameter for compaction quality. Sr in-place is usually determined with sand cone method. The method is time consuming as well as prone to error due to potential water seepage and deformation of soft soils. With its ability to provide remote and continuous monitoring in real time, Time Domain Reflectometry (TDR) has the potential to serve as a viable alternative method for determination Sr in-place. This research presents models developed using different calibration methods for the prediction of Sr with TDR measured dielectric permittivity. The accuracy of the models was evaluated with cross-validation. The results show that TDR is able to predict the Sr with less than ±5% deviation for 93% of the specimens in the validation set within the saturation ratio range of 0.01–0.93 cm3 cm−3 with an appropriate model.