Research On Numerical-cum-Experimental Method For Determining Thermal Conductivity Of Pavement Materials And Its Application

Thermal conductivity is an important parameter for studying pavement heat transfer behavior,which is critical for improving the accuracy of pavement temperature field prediction and optimizing pavement structure design.Existing finite element models for predicting the thermal conductivity of pavement materials need to reconstruct the complex microstructures,which are difficult to operate and calculate inefficiently.The experimental method requires specialized equipment to determine the thermal conductivity of the pavement materials,and it must be repeated whenever the internal composition of the pavement material changes in proportion.Based on Fourier heat transfer theory,this paper employs finite element simulation methods to develop 2D and 3D row-column finite element models with regular forms to predict thermal conductivity,and it combines genetic algorithms to propose a new method based on numerical-cum-experimental to inversely analyze the thermal conductivity of pavement material components and predict the thermal conductivity of pavement materials with different composition proportions.And a cooling model of newly laid asphalt pavement was also developed to study the pavement’s cooling behavior under complex conditions,and the more reasonable pavement opening criteria was proposed.In this paper,a 2D row-column finite element model with regular form is developed to predict the thermal conductivity of pavement materials based on the volumetric parameters of each component in asphalt mixture and cement concrete.The characteristics of previous finite element models were evaluated,and it was found that the shape and structure of the pavement material components have little influence on the prediction thermal conductivity,but the key is the random distribution of its components.A simplified row-column finite element model is proposed,and a convergence analysis is carried out.By comparing to the existing numerical simulation model and the measured results,the feasibility and validity of the 2D row-column finite element model are validated.It is shown that the 2D row-column finite element model with regular form has the same prediction accuracy with complex microstructure model;the proposed model simplifies the construction process of the numerical model and improves the computational efficiency.The 2D row-column finite element model is extended to create a 3D model,the convergence of the three-dimensional model is analyzed,and the 3D model is validated using experimental data.The effects of heat transfer dimension,distribution mode,air content,and thermal conductivity of the binder on the pavement material are investigated,as well as the advantages of each of the two-dimensional and three-dimensional models.The result shows that there are more heat transfer channels in three dimensions,with slightly higher prediction accuracy than in two dimensions,and the difference is within 5%.The mode of distribution of the internal substances has a particularly strong influence on the overall thermal conductivity of pavement materials,with results for the random distribution mode falling somewhere between the series and parallel modes.The effect of air content on asphalt mixture thermal conductivity is greater than cement concrete,and asphalt mixture thermal conductivity decreases continuously as the void content increases.As the thermal conductivity of the binder increases,the thermal conductivity of the pavement material increases to various levels,and this change has a more significant effect on the concrete.The 3D model has a higher accuracy,while the 2D model has a lower computational cost;both can be used to predict the thermal conductivity of pavement materials.To improve model processing efficiency,ABAQUS secondary development technology was used to automate modeling and post-processing.The framework for coupling genetic algorithms with ABAQUS was developed;and the numerical-cum-experimental approach was proposed to determine the thermal conductivity of pavement materials and its compositions.The steady-state experiments were designed and validated for the method.The method,when combined with 2D and 3D row-column finite element models,can predict the thermal conductivity of pavement materials at different volume contents,reducing the number of experimental tests and saving time and material.To investigate the cooling behavior of newly laid asphalt pavement under complex working conditions,a cooling model based on thermodynamic theory and the finite element method was developed.Asphalt pavements are more susceptible to temperature effects than concrete pavements,so this paper focuses on the temperature field of asphalt pavements.This paper establishes 216 finite element models for newly laid asphalt pavement based on the pavement maintenance conditions of Singapore Changi Airport,and studies the cooling behaviours of airport pavement under different weather conditions,wind speeds,and paving strategies to meet the demand of repaired asphalt pavement opening traffic under complex working conditions.The study shows that the surface temperature of newly laid asphalt pavement is not suitable as the pavement opening criteria,when the road surface temperature cools to 60℃,the internal temperature of the newly paved asphalt layer is still at a high level,and it is recommended to choose the position of 1/2,2/3 or 3/4 of the newly paved layer as the reference point of the pavement opening temperature.