Abstract
The objective of this study is to investigate the impact of aggregate gradation and Superpave volumetrics on thermal properties of asphalt mixtures. Thirty asphalt concrete specimens with three different Nominal Maximum Aggregate Sizes (NMAS) of 19.0, 12.5 and 9.5 mm and two levels of gradation coarseness: fine gradation (FG) and coarse gradation (CG) were prepared. The Transient Plane Source (TPS) method was used to determine thermal properties. Based on the analyses performed, it was concluded visually that heat transfer is highly dependent on the contact area which, in turn, is related to air void volume. Unlike NMAS, the gradation coarseness (G) has some effect on thermal conductivity (TC) (p-value symbolscript 0.008 < 0.05). On the contrary, the NMAS, unlike gradation coarseness, has a significant effect on thermal diffusivity (TD) with a p -value of 0.025. Both gradation coarseness and NMAS have a significant effect on thermal diffu-sivity (p-value symbolscript 0.029 < 0.05) and do not have much effect on specific heat (SH) where p-value is greater than 0.05. Nevertheless, it was noticed that the thermal diffusivity increased linearly with an increase in aggregate size from 85.64 mm2/s to 125.79 mm2/s at 9.5 NMAS FG and 19 mm NMAS FG, respectively. Whereas the specific heat and thermal conductivity increase as the asphalt content (AC) increases. Consequently, the highest average values of thermal conductivity and specific heat are 2 W/m.K and 0.025 M.J/m3. K, respectively, at the highest asphalt content of 6.1%. Multiple linear regression, non-linear regression and deep learning (MLR, NLR and DL) equations were developed. The non-linear regression resulted relatively in the best predictive power of all equations. Finally, larger datasets are needed to predict thermal properties with higher confidence.