Laboratory determination of thermal conductivity is a commonly used method for analyzing problems related to thermal behaviors for different applications. However, laboratory determination of thermal conductivity of asphalt paving mixtures has remained a great challenge because it requires special devices and setups that are commonly not available in standard pavement engineering laboratories. Moreover, the changes in the thermal conductivity of the asphalt mixtures with a reduction in the air void content require repeated determination of thermal conductivity, which is time-consuming and economically unviable.
To address the above challenges, a team of Chang’an University researchers: Dr. Chu Longjia, Mr. He Lei (PhD candidate) and Professor Dr. T. F. Fwa developed a new computer-based numerical model to determine the thermal conductivity of asphalt paving mixtures. In particular, a finite element method was used to determine the thermal conductivity of asphalt mixtures with known volume proportions of the mixture constituents: air, asphalt binder, and aggregate. Their work is currently published in the research journal, Construction and Building Materials.
In their approach, the asphalt mixture to be analyzed was represented by a two-dimensional row-column structure. Based on the row-column structure, a computer simulation model of heat conduction in an asphalt mixture was developed to derive the thermal conductivities of the mix. At this juncture, two cases were considered. The first case involved simulations with known thermal conductivity values of the mixture constituents, while the second case involved simulations with unknown thermal conductivities of one or more of the constituents. For the second case, the authors calibrated the model using a genetic-algorithm to determine all the unknown thermal conductivities from the known laboratory or field values of mixture k. Lastly, the feasibility of the model in determining the thermal conductivities of asphalt mixtures was demonstrated using examples.
The authors observed that for known thermal conductivity values of the asphalt binder and aggregates, the thermal conductivity of the asphalt mixture was directly given by the converged solution of the simulation analysis. However, in cases where one or more of the constituent thermal conductivities were unknown, the model was calibrated to determine the required thermal conductivity values from the existing laboratory or field values of mixture k. Once calibrated, the proposed model was used to determine the thermal conductivity of the asphalt mixture at different air void contents without the need for further field or laboratory measurements. Furthermore, the row-column finite element model was more efficient in generating random structures of asphalt mixtures than microstructure simulation methods. It also produced predicted thermal conductivity values of asphalt mixture with the same accuracy level as those produced by sophisticated microstructure simulation methods.
In summary, the study presented a two-dimensional row-column finite element model to determine the thermal conductivity of asphalt mixtures. Results showed that the model could be applied directly to determine the thermal conductivity of an asphalt mixture when the constituent thermal conductivity values of air, binder, and aggregate are known. Moreover, in cases where the constituent thermal conductivities are unknown, a calibration procedure coupled with the presented numerical model can be applied to obtain the thermal conductivity of the mixture. In a statement to Advances in Engineering, Professor T. Fwa, the lead author said that their simple model could be potentially used in different practical applications and other similar form of composite materials.
Chu, L., He, L., & Fwa, T. (2020). Determination of thermal conductivity of asphalt paving mixtures using finite element method. Construction and Building Materials, 243, 118250.