Density Measurement of Compacted Asphalt Mixtures Using Non-Destructive Ground Penetrating Radar

This thesis describes the development of Ground Penetrating Radar (GPR) system based on the electromagnetic wave reflection to determine the density of road pavement. The proposed method is simple, fast, non-destructive and within an acceptable accuracy of road pavement density. The theoretical analysis based on the three existing GPR Mixture Model (GMM) methods has been improved to produce the most optimized function to be incorporated within the proposed GPR system. The study involves three main procedures which are theoretical analysis, laboratory scale experimentation and reliability analysis. From these studies, the Lichtenecker Mixture Model is found to be the most accurate function compared to the other models like Nelson and Landau due to the smallest mean error between the prediction and the experimental result. During the laboratory experimentation, an engineering GPR prototype has been developed and used to measure the road pavement density of the road pavement slab sample. The GPR system consists of the transmitter which is signal generator as a microwave source, horn antenna for transmitting and receiving the signal, directional coupler with an adapter and spectrum analyzer to analyze the received signal. Nine road pavement slabs of middle boundary and ten slabs of upper and lower boundary of Hot Mix Asphalt (HMA) gradation were developed and tested at four different frequencies within the range of 1.7-2.6 GHz. The predicted signal attenuation from the theoretical analysis is compared to the signal attenuation measured from the laboratory experimentation. The comparison produces the relative error between these two results and it is used in the optimization process. The finding from the optimization process suggested that three additional constant parameters which are Volume factor, Permittivity factor and Attenuation factor need to be included to improve the existing GMM model. A field test had been conducted as an outdoor reliability analysis to validate the optimized GMM model. From the field test, it shows that the proposed GPR system works well with an error range from 3.37 % to 4.72 % for nine locations. Finally, a complete GPR system has been developed based on the optimized GMM attenuation curve to predict the density of a real road pavement

Preview PDF FK_2009_10a.pdf Download (545kB)

Actions (login required)

View Item