The feasibility of DEM to analyze the temperature field of asphalt mixture

https://doi.org/10.1016/j.conbuildmat.2015.12.192Get rights and content

Highlights

  • Digital sample was reconstructed considering discontinuity of asphalt mixture meso-structures.

  • A discrete element model was established considering the contact thermal resistance.

  • Numerical simulation results from DEM model were validated with experimental results.

  • A better heating scheme was proposed to improve bulk temperature of asphalt mixture.

Abstract

To discuss the feasibility of the discrete element method (DEM) to analyze the temperature field of asphalt mixture materials, the digital specimen of rutting plate was reconstructed based on the mix design of asphalt mixture. Also, considering the contact thermal resistance, a discrete element thermal analysis model was established based on the discrete element heat conduction equation and its discretization. The temperature field distribution in digital specimen was discussed using the DEM model, and which was compared with the experimental results on the actual rutting plate. The results indicated that the discontinuity of mesoscopic structures in asphalt mixture was fully considered in digital specimen based on the digital image processing. The internal structures of asphalt mixture specimen were reappeared, and the transient heat conduction discrete element model was established after the heat flux was applied on the specimen surface. Further, thermal responses of the upper part were faster than that of the lower part in discrete element model during continuous heating. The bulk temperature in digital specimen continued to rise after the heating was over. Similarly, the measured temperatures of upper part were higher than that that of lower part in rutting plate under both continuous and intermittent heating schemes. However, the surface temperature was lower under the intermittent heating method, reducing the asphalt aging and improving the bulk temperature and its uniformity of the specimen. It is concluded that numerical simulation results using DEM are basically consistent to experimental results. The DEM may become a new promising approach to study the temperature field distribution and other properties of asphalt mixture materials.

Introduction

Asphalt mixture is a kind of important paving materials, and its properties are closely related with temperature, internal mesoscopic structures [1]. It is necessary to understand the temperature field of asphalt mixture or asphalt pavement because their properties are sensitive to temperature. The numerical simulation has become more and more important along with the advance in computer technology and the development in calculating method [2]. Currently, the numerical simulation method of asphalt mixture temperature field mainly includes finite element method (FEM) and discrete element method (DEM) [3].

One major drawback of the FEM is that continuum-based methods are unable to include the stochastically distributed microscopic effects in the macroscopically oriented calculations [4]. It is difficult for FEM to consider the mesostructure non-continuity in asphalt mixture. On the other hand, the DEM is one method with which these effects can be considered. It is possible to make continuum-based calculations not only in the mechanical field but also in the thermal field using DEM. Therefore, the DEM becomes one of the important methods to analyze the different performance of asphalt mixture.

Zhang et al. [5] studied the relationship between effective thermal conductivity and mechanical properties of granular materials using the DEM. Tsory et al. [6] reported the roughness effects of granular materials on the heat transfer by means of DEM and fluid mechanics. Gui et al. [7] analyzed the thermal conduction of granular materials in the mixed state and heat transfer among the particles in a rotating drum based on the DEM.

Zhou et al. [8] proposed to investigate the anisotropic thermal conductivity of granular materials using homogenization technique, and to validate the effectiveness of that method by means of numerical simulation based on DEM. Liu et al. [9] further investigated the viscoelastic model of asphalt mixture using the DEM, and explained the contact model between asphalt mortar and coarse aggregate in detail.

Additionally, a model was established to discuss the thermal conduction and heat transfer among particles, and the temperature field of particles was analyzed through the DEM by Shimizu [10]. The temperature field variation of alumina particles was discussed based on the DEM, and the residual stress due to thermal expansion was calculated by Nohut [11]. Pennec et al. [12] investigated the thermal conductivity of biomass particle materials using the two methods of DEM and FEM, and the numerical simulation results were compared with the experimental results to prove the reliability of numerical methods.

Recently, Van Lew et al. [13] discussed the thermal effects of pebble failure in an ensemble of lithium ceramic spheres on the basis of DEM. Hahn et al. [4] demonstrated the reliability of the DEM to analyze the temperature field of granular materials, and pointed out the method for determining the material parameters. Dondi et al. [14] discussed the effect of aggregate shape and angularity on the properties of asphalt mixture by means of the DEM to improve the performance of asphalt mixture.

Chen et al. [15] evaluated the aggregate skeleton structure characteristics in asphalt mixture and their resistance to deformation at high temperature using the DEM. The viscoelastic behavior of asphalt mixture was studied in terms of the DEM, and the dynamic properties of asphalt mixture were captured through implementing Burger’s contact model by Feng et al. [16]. Liu et al. [17] found that the element size of the discrete element model of asphalt mixture had obvious influence on the computing time step, the computing time and the mechanical properties. Obviously, the DEM plays a more and more important role in the performance research of asphalt mixture, and has gradually become one of important methods for numerical simulation [18].

However, the current DEM was mostly utilized to analyze gradation composition, macroscopic deformation, and mechanical properties of asphalt mixture under a constant temperature conditions through the simulation of fatigue test, uniaxial compression test and rutting test. Few experiments were conducted to validate the feasibility and reliability of the DEM which was used to analyze the temperature field of asphalt mixture. In addition, the current DEM usually considered indirectly the effects of temperature on the properties of asphalt mixture through the time temperature equivalent principle, which lead to a deviation with actual conditions. The boundary heat flux was seldom applied directly on the model to analyze properties of asphalt mixture materials [19].

In this study, the discontinuity of mesoscopic structures in asphalt mixture was fully considered when the digital specimen was reconstructed according to the cross section image of rutting plate based on the digital image processing technique and the DEM. Considering the contact thermal resistance, a discrete element thermal analysis model was established to analyze the temperature field distribution in asphalt mixture specimen. Then the temperature field distribution in digital specimen was discussed using the DEM model to simulate the heating process of hot in-place recycling of asphalt pavement. Further, numerical simulation results were compared with the experimental results measured on the rutting plate under the same heating mode and boundary conditions as those in discrete element model. This may develop a new promising method to study the temperature field distribution and various properties of asphalt mixture.

Section snippets

Heat conduction equation and its discretization

Thermal module in PFC 2D can be used to simulate transient heat conduction problems, in which the particle is seen as a thermal memory, and the contact between particles forms the heat pipe [20]. Each particle is looked as a hot memory in the process of heat conduction, but only when the particles contact mutually or form a connection, the heat pipe is activated. The heat is transferred in the activated heat pipes [20].

In this study, asphalt mixture is assumed to be the heat isotropic material,

Reconstruction of digital specimen of asphalt mixture

In view of the irregularity of aggregate particle distribution in asphalt mixture specimen, the internal mesoscopic structures of asphalt mixture are reappeared based on the mix design of the actual rutting plate. The rutting plate was prepared using SMA-13 asphalt mixture whose final combined aggregate gradation is shown in Fig. 1, and the asphalt content is 5.7 %. The technical indicators of raw materials such as SBS modified asphalt, aggregate, filler, stabilizer, etc. meet the relevant

Heating experiments on asphalt mixture specimen

To validate the feasibility of the DEM for analyzing the temperature field of asphalt mixture materials, the temperature field distribution at different depths in the rutting plate were measured using the heating simulation experiment under the same boundary conditions as that in the DEM model.

Conclusions and recommendations

The temperature field of asphalt mixture materials is analyzed using the DEM, and the numerical results are compared with the experimental results. The main conclusions are summarized as follow.

  • (1)

    The digital specimen of asphalt mixture was reconstructed, and the discontinuity of internal mesoscopic structures was reappeared. Considering the contact thermal resistance, the transient heat conduction discrete element model was established based on the heat conduction equation and its discretization

Acknowledgements

Authors would like to thank the financial support from National Natural Science Foundation of China (No. 51378264) and Provincial Natural Science Foundation of Jiangsu (No. BK20141475) and Jiangsu Provincial Department of Education for the Qing Lan Project (No. 2014-23) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). Also we would like to thank Advanced Analysis & Testing Center of Nanjing Forestry University for the assistance in experiments.

References (25)

Cited by (0)

View full text