Modal Analysis Software Applications to Characterize the Complex Modulus of Asphalt Concrete
Anders GudmarssonShow Abstract
The time and temperature dependent dynamic modulus of asphalt concrete is a key parameter to modern pavement design and to pavement quality. However, accurate measurements of the complex modulus through conventional cyclic testing are expensive regarding both time and costs. Novel test methods based on modal analysis facilitate economic, accurate and faster determination of linear viscoelastic material properties of asphalt concrete. The modal analysis is performed by using an impact hammer and an accelerometer to measure frequency response functions of asphalt concrete specimens with free boundary conditions. The complex modulus and complex Poisson’s ratio are determined by matching finite element computed frequency response functions to the measurements. This paper present one software application to perform the modal testing and one application to compute and optimize frequency response functions to determine the complex modulus and complex Poisson’s ratio. The software applications greatly simplifies the methodology to determine material properties through modal analysis of asphalt concrete specimens.
Rheological Characterization of Asphalt Fine Aggregate Matrix Using Dynamic Shear Rheometer
Zejiao Dong, Harbin Institute of TechnologyShow Abstract
Xiangbing Gong, Harbin Institute of Technology
Zhiyang Liu, Harbin Institute of Technology
Asphalt fine aggregate matrix (FAM) is a decisive component directly related to the trans-scale mechanism of asphalt mixture, thus it is necessary to investigate material properties of FAM. Prior to preparing FAM specimens, and the asphalt content was calculated by keeping filler-bitumen (FB) ratio the same as it in corresponding mixtures. A non-destructive fabrication method was developed to compact FAM cylinders, and the joint base was designed to be concentric with the loading axis of testing system. Dynamical mechanical responses of FAM were studied using the Dynamic Shear Rheometer (DSR). Two repeated tests prove that the FAM compactor and the jointed base meet the requirement of data validation. Results show that rheological performances of FAM are significantly affected by asphalt content, gradation, air void content, and testing frequency. Air voids are concluded that the stability of FAM, and the fiber is demonstrated to improve stability of FAM-F.
Can Asphaltic Concrete Permanent Deformation Test Methods Be Used as Surrogates for Each Other?
Yasir Ali, University of QueenslandShow Abstract
Sarfraz Ahmed, National University of Sciences and Technology
Rutting or permanent deformation is one of the severe distresses manifested in flexible pavements caused by the densification or lateral plastic flow of materials under repeated heavy traffic loads. Various laboratory tests have been devised to determine the rutting propensity and optimize field performance of hot mix asphalt (HMA) mixtures as a part of the HMA mix- and structural- design processes. Cognizant of widely used approaches to predict permanent deformation in flexible pavements, this study attempts to develop a relationship of flow number (FN), dynamic modulus (DM), and uniaxial repeated load permanent deformation (RLPD), and establishes the fact if they could be surrogate to each other. Also, a novel parameter – FN Index is explored and used to determine the rutting potential of HMA mixtures. Twelve (12) globally practiced HMA mixtures are investigated with three different performance grade binders and single limestone aggregate source. Superpave gyratory compacted specimens are subjected to performance testing (FN, DM, and RLPD) and results indicate a strong correlation of FN Index with DM and exhibited better correspondence than traditional FN (cycles) parameter.
Development of Duplicate Shear Test for Asphalt Mixtures
Mohammadreza Khajeh Hosseini, University of Texas, ArlingtonShow Abstract
Stefan Romanoschi, University of Texas, Arlington
Nickey Akbariyeh, University of Texas, Arlington
Reza Saeedzadeh, University of Texas, Arlington
Permanent shear deformation is the main cause for rutting in asphalt concrete layers. The paper presents the development of a new test device to replicate similar loading conditions and constraints of the Superpave Shear Tester (SST). The Duplicate Shear Tester (DST) determines the mechanical average of the shear properties and responses of two specimens loaded simultaneously. The DST device is simple and inexpensive and can be used with any universal testing machine that can provide repeated dynamic and static axial loads in a temperature-controlled environment. The two most used SST tests, the Frequency Sweep Test at Constant Height (FSCH) and the Repeated Shear Test at Constant Height (RSCH) were performed with the DST device mounted inside of a UTM-25 test system.
To evaluate the effectiveness of the DST, four asphalt surface mixtures were subjected to FSCH tests at 30°C and RSCH tests at 50°C. For each mixture, three pairs of specimens were tested. The data analysis indicated that the DST test is highly repetitive and reliable in measuring the shear dynamic modulus and phase angle at load frequencies between 0.5 Hz to 10 Hz, for which a variability of less than 10% was observed. A variability higher than 10 percent was recorded for the FSCH tests done at frequencies lower than 0.5 Hz. A variability slightly higher than 10 percent was observed in the RSCH test for the permanent shear deformation for two of the four asphalt mixtures.
Research on Local Deformation Property of Asphalt Mixture Using Digital Image Correlation
Chao Xing, Harbin Institute of TechnologyShow Abstract
Yiqiu Tan, Harbin Institute of Technology
Xueyan Liu, Delft University of Technology
Kumar Anupam, Delft University of Technology
Tom Scarpas, Delft University of Technology
Failure behavior of asphalt mixture is the process of local deformation accumulation, therefore, local deformation investigation is the base work of failure behavior. Digital image correlation (DIC) method is an optical method which can calculate the local deformation in the failure process. This paper aim to evaluate the precision and select the optimal test parameter for DIC deformation measurement based on natural speckle image, and investigate the local deformation characteristic in the failure process for different mixtures. The results show the image quality of AC with smaller normal maximum aggregate size (NMAS) is better for DIC deformation calculation, the relative deviation of virtual translation and shear test can satisfy the precision requirement, the optimal subset size is 31. Based on the strain distribution, the optimal local deformation measure length for indirect tensile test is set as 50mm, the critical steady strain is proposed based on R 2 obtained by the least squares fitting method, according to local deformation characteristic, the results show SMA can bear more deformation before unstable deformation stage and failure.
Viscoelastic-Plastic-Fracture Modeling of Asphalt Mixtures Under Monotonic and Repeated Loads
Yuqing Zhang, Aston UniversityShow Abstract
Fan Gu, Auburn University
Bjorn Birgisson, Texas A&M University
Robert Lytton, Texas A&M University, College Station
Rutting and cracking occur simultaneously in asphalt mixtures as observed in the field and laboratory. Existing mechanical models have not properly characterised the viscoelastic and viscoplastic deformation together with cracking due to model deficiencies, parameter calibration and numerical inefficiency. This study develops viscoelastic-plastic-fracture (VEPF) models characterising viscoelasticity (VE) by Prony model and viscoplasticity (VP) by Perzyna’s flow rule with a generalized Drucker-Prager (GD-P) yield surface and a non-associated plastic potential. Viscofracture (VF) cracking is modelled by a viscoelastic Griffith criterion and a pseudo J-integral Paris’ law for crack initiation and propagation, respectively. The VEPF models are implemented in a finite element (FE) program using a weak form partial differential equation (PDE) modelling technique with no need to program user defined material subroutines. Model parameters were derived from the material fundamental properties determined by dynamic modulus, strength and repeated load tests. Simulations indicate that the VE-VP-VF characteristics are effectively modelled by the VEPF models for different asphalt mixtures and at different confinements and temperatures. An asphalt mixture under monotonic compressive loads exhibits a sequenced process including a pure VE deformation stage, a coupled VE-VP deformation stage, a VE-VP deformation coupled with VF initiation and propagation stage, and then a VE-VF rupture stage with saturated VP deformation. The asphalt mixture under repeated loads yields an increasing VP strain at an increasing rate during the first half of the haversine load while a decreasing VP strain increment with load cycles. The PDE-based FE is demonstrated to be capable of effectively modelling the coupled VE-VP-VF behaviours of the asphalt mixtures.
Quantification of Inherent Uncertainty in Relaxation Modulus and Creep Compliance of Asphalt Mixes
Hussein Kassem, Beirut Arab UniversityShow Abstract
Ghassan Chehab, American University of Beirut
Shadi Najjar, American University of Beirut
Advanced material characterization of asphalt concrete is essential for realistic and accurate performance prediction of flexible pavements. However, such characterization requires rigorous testing regimes that involve mechanical testing of a large number of laboratory samples at various conditions and set-ups. Advanced measurement instrumentation in addition to meticulous and accurate data analysis and analytical representation are also of high importance. Such steps as well as the heterogeneous nature of asphalt concrete (AC) constitute major factors of inherent variability. Thus, it is imperative to model and quantify the variability of the needed asphalt material’s properties, mainly the linear viscoelastic response functions such as: relaxation modulus, E(t), and creep compliance, D(t). The objective of this paper is to characterize the inherent uncertainty of both E(t) and D(t) over the time domain of the their mastercurves. This is achieved through a probabilistic framework using Monte Carlo simulations and First Order approximations, utilizing E* data for six AC mixes with at least eight replicates per mix. The study shows that the inherent variability, presented by the coefficient of variation (COV), in E(t) and D(t) is low at small reduced times, then increases with the increase in reduced time. At small reduced times, the COV in E(t) and D(t) are the same; however, differences become significant at large reduced times. Additionally, the distributions and COVs of E(t) and D(t) are mix-dependent. Finally, a case study is considered in which the inherent uncertainty in D(t) is forward propagated to assess the effect of variability on the predicted number of cycles to fatigue failure of an asphalt mix.
Generalized Regression Approach to Develop Predictive Models for Dynamic Modulus and Phase Angle of Asphalt Mixtures
RASOOL NEMATI, AECOMShow Abstract
Eshan Dave, University of New Hampshire
Dynamic modulus (|E*|) and phase angle (δ) are viscoelastic properties of asphalt mixtures that are necessary for determination of constitutive response of mixtures to traffic and thermal loading. These parameters are necessary for analysis and design of asphalt pavements. While a number of |E*| and δ predictive models have been developed, many of them require lab measured properties, such as, gradation and binder complex modulus (G*) for accurate predictions. Furthermore, the majority of previous work has been focused on prediction of |E*| and only few models have been explored for prediction of δ. This research utilized generalized regression modelling to develop |E*| and δ prediction models using nominal asphalt mix properties (such as, asphalt content, air void and aggregate size) that are often readily available during initial mixture design and specification process. Using nominal properties not only eliminates need for even the simplest lab tests, but also provides the pavement design engineers with the ability to conduct mechanistic analysis with reliable material properties. A total of 81 asphalt mixtures and 4374 lab measurements of |E*| and δ each were used for model development. The accuracy of prediction models is verified through statistical analysis. A case study is presented to demonstrate applicability of these models for performance predictions. The case study also presents verification that there are minimal differences in performance predictions when using predicted properties as compared to lab measured properties.
Time-Temperature-Aging-Depth Shift Functions for Dynamic Modulus of Asphalt Mixtures
Meng Ling, Texas A&M UniversityShow Abstract
Xue Luo, Zhejiang University
Fan Gu, Auburn University
Robert Lytton, Texas A&M University, College Station
This study aimed to develop a new aging model for predicting the dynamic modulus of laboratory-aged and field-aged asphalt mixtures. The tensile creep test and direct tension test were conducted to measure the tensile dynamic moduli of 12 laboratory-aged and 16 field-aged asphalt mixtures, respectively. The modified CAM model was used to construct the dynamic modulus master curves for these mixtures. A time-temperature-aging superposition concept was employed to characterize the viscoelastic properties of laboratory-aged and field-aged asphalt mixtures. Unlike the time-temperature superposition, the time-temperature-aging superposition allowed to shift the dynamic modulus master curves of asphalt mixtures at different aging times to a reference aging time. The aging shift factors were used to evaluate the effect of oxidative aging on the viscoelastic behavior of asphalt mixtures. The final master curve of laboratory-conditioned mixtures was constructed including the effect of temperature, loading frequency and aging time. Different from the laboratory-aged mixtures, the dynamic modulus curves of field-aged asphalt mixtures change with the pavement depth. Therefore, an additional depth shift factor was developed to construct the final master curves of field-aged asphalt mixtures. The final master curve was capable of predicting the viscoelastic behavior of field-aged asphalt mixtures at different temperature, loading frequency, aging time, and pavement depth.
Generalization of the Hirsch Dynamic Modulus Prediction Model for Asphalt Mixtures
Cheng Zhang, Tongji UniversityShow Abstract
Shihui Shen, Tongji University
Jonathan Jia, Jiaxing Jiahai Construction Co., Ltd.
Shuai Yu, Pennsylvania State University
Dynamic modulus of asphalt mixture plays a crucial role in the material characterization, pavement design, and performance prediction of asphalt pavement. It is, however, costly and time-consuming to be measured directly in the laboratory. Significant effort has been spent on the development of dynamic prediction models. This paper aims to modify one of the most popularly used dynamic modulus prediction models, Hirsch model, by replacing its constants with more fundamental material property models based on the theory of elasticity and viscoelasticity. Such effort will allow the model to be more versatile to a variety of newer mixture types while maintain the simplicity of the Hirsch model at the same time. A total of 26 asphalt mixtures sourced from China and the USA were utilized in this study to evaluate the predictive quality of the generalized Hirsch model, as compared to the 2006 Witczak model and the original Hirsch model. It was found that the generalized Hirsch model produced the best predictive quality among the three models. In addition, the generalized model was shown to be suitable for predicting the dynamic modulus of asphalt mixtures with high reclaimed asphalt pavement (RAP) contents.
Water Vapor Diffusion in Asphalt Mixtures Under Various Relative Humidity Differentials
Tingting Huang, Wuhan UniversityShow Abstract
Rong LUO, Wuhan University
Derun Zhang, Wuhan University
Robert Lytton, Texas A&M University, College Station
It has been demonstrated that water vapor consistently transports in asphalt pavements. The relative humidity differential, which exists between the subgrade below and the atmosphere above the pavement, is a major contributor to the water vapor diffusion. However, the effect of the relative humidity differential has not been quantified on water vapor diffusion in asphalt mixtures. This paper designed a laboratory experiment to investigate the water vapor diffusion in asphalt mixtures under a number of relative humidity differentials. The designed experiment consisted of five individual water vapor diffusion tests, each of which was performed at a specific relative humidity differential in the pre-vacuumed measuring cell of the test equipment. Two diffusion models were developed based on the Fick’s second law of diffusion. It was found that the two-dimensional model provided an accurate characterization of the water vapor diffusion in both radial and axial directions of the cylindrical test specimens.
A linear model was established between the relative humidity differential and the total diffused mass based on the real gas law. An inverse proportionality was identified between the relative humidity differential and the diffusivity in both radial and axial directions. The determined diffusivity values were further converted to the corresponding diffusivity values under 1 atmosphere, which were in agreement with the diffusivity data reported in the literature. A linear model was derived for the relationship between the relative humidity differential and the diffusivity under 1 atmosphere, which indicated faster water vapor diffusion in asphalt pavements when subjected to a larger relative humidity differential between the subgrade and the atmosphere. The diffusivity in the radial direction was always larger than that in the axial direction despite the variation of the relative humidity differential. This fact implied the anisotropic distributions of air voids and aggregates in the asphalt mixture specimens, which assisted water vapor in diffusing more easily in the radial direction than in the axial direction.
Characterization of Linear Viscoelastic, Nonlinear Viscoelastic, and Damage Stages of Asphalt Mixtures
Hanqi Liu, Wuhan UniversityShow Abstract
Rong LUO, Wuhan University
Yuqing Zhang, Aston University
It has been demonstrated that asphalt mixtures experienced linear viscoelastic stage, nonlinear viscoelastic stage and damage stage when subjected to controlled-strain repeated direct-tension (RDT) tests with increasing strain levels. However, the linear viscoelastic properties of asphalt mixtures are usually muddled up with their nonlinear viscoelastic properties. These confusions directly lead to the incorrect determination of the pseudostrains and dissipated pseudostrain energies (DPSEs) in the nonlinear viscoelastic stage and damage stage. This study investigated the material properties of fine aggregate mixture (FAM) specimens in all three stages. These three stages were differentiated and characterized in terms of the viscoelastic stress, pseudostrain and DPSE. The definitions of viscoelastic stress, reference modulus and pseudostrain were rigorously established to assure that the material properties in the linear viscoelastic stage were the reference properties and that the sole linear viscoelastic effect was eliminated when determining the pseudostrain and DPSE in the three stages. The characteristics of the DPSE in the three stages were found to be: 1) the DPSE of any loading cycle was zero in the linear viscoelastic stage; 2) in the nonlinear viscoelastic stage, the DPSE of each loading cycle remained approximately the same with the growth of the number of loading cycles, and the DPSE increased to a larger value when the strain level of the RDT test increased to a higher level; 3) in the damage stage, the DPSE of the loading cycle increased as the number of loading cycles increased. This study strictly distinguished the linear viscoelasticity from the nonlinear viscoelasticity of the asphalt mixtures, which is critical for the accurate determination of the DPSE spent in overcoming the nonlinear viscoelasticity and in developing damages, such as cracking and permanent deformation, in the asphalt mixtures.
Identification of Potential Location for Debonding Between Asphalt Layers Under a Moving Load Using Viscoelastic and Equivalent Elastic Approaches
Bongsuk Park, University of FloridaShow Abstract
Jian Zou, University of Florida
Reynaldo Roque, University of Florida
David Hernando, University of Florida
Jeremy Waisome, University of Florida
Poor bonding conditions between asphalt layers may lead to two major distresses: slippage and cracking. Whereas slippage may result from a single high load event, cracking may involve progressive breakdown of the interface under repeated traffic loads. Field observations appeared to associate near-surface longitudinal cracking in the wheelpath with localized interface debonding. This study focused on identifying the potential location and extent of debonding. This effort will pave the way for achieving a better understanding of how localized interface debonding may initiate longitudinal cracks that propagate to the surface. 3-D asphalt pavement models were developed using ABAQUS. Effects of viscoelastic behavior of asphalt materials on responses of pavements in relation to debonding when subjected to a wheel load moving at varying speeds were evaluated to better simulate field conditions. Moreover, feasibility of an equivalent elastic approach was investigated. The results showed that 3-D equivalent elastic analysis appeared to be sufficient to accurately determine the potential location and extent of localized debonding. It was found that localized debonding can potentially occur in a 10- to 14-inch wide strip around the mid-depth of AC layer below the wheelpath. It was noted that restraint at the edge of the de-bonded strip on the transverse cross-section may lead to stress concentrations that may initiate longitudinal cracks. Since no such restraint occurs along the longitudinal cross-section, it was recommended that further analyses regarding assessing impact of localized debonding and identifying potential link to longitudinal cracking may be properly conducted as a plane-strain problem.
Analysis of Flow Number Test Data on Asphalt Mixtures from Michigan for Use in Pavement Mechanistic-Empirical Software
M. Emin Kutay, Michigan State UniversityShow Abstract
Anas Jamrah, Marathon Petroleum Company LP
Hilmi Bayraktaroglu, Middle East Technical University
Rutting is one of the major distress types in flexible pavements. This type of distress results from a combination of densification, viscoplastic and shear deformation in one or more pavement layers. The flow number (FN) test is known to be a good indicator of rutting susceptibility of Asphalt Concrete (AC). The FN test results can also be used to calibrate the rutting model coefficients of the AASHTOWare Pavement ME software. The objectives of this study were: (i) to investigate the rutting characteristics of asphalt mixtures commonly used in Michigan, (ii) obtain local calibration coefficients ( br1, b r2 and b r3) for the rutting model used in Pavement ME, and (iii) make recommendations on the use of these factors.
Permanent Deformation Characterization of Asphalt Mixtures Using Minimum Strain Rate, LVECD Program, and Triaxial Stress Sweep Test
Dahae Kim, South Carolina Department of TransportationShow Abstract
Youngsoo Kim, North Carolina State University
The rutting resistance of asphalt concrete typically is assessed using flow number tests in the lab in accordance with AASHTO TP 79. However, the flow number represents the rutting resistance of the material in terms of ranking for only a specific test condition. A significant amount of testing effort is needed to evaluate rutting resistance using the flow number test under various loading conditions and temperatures. Therefore, researchers have developed alternative test methods to reduce the testing effort. For example, the incremental repeated load permanent deformation (iRLPD) test and the triaxial stress sweep (TSS) test are two promising protocols that can predict the permanent deformation of asphalt mixtures efficiently and accurately.
This study compares the minimum strain rates (MSRs) obtained from iRLPD and TSS tests to examine the ability of the TSS test to analyze the MSRs of asphalt mixtures. In addition, the viscoplastic shift model calibrated from the TSS test that is implemented in the Layered ViscoElastic pavement analysis for Critical Distresses (LVECD) program was used to predict the rut depths of 16 pavement sections from four different sources. The MSRs and rut depths predicted from the LVECD program were compared to actual measured rut depths. The results of this study indicate that the MSR results can be used to rank the rutting susceptibility of only single layer asphalt pavements. For a multilayered pavement system, structural level analysis is necessary to predict the accurate rutting performance and rut depths of the test sections.
Dynamic Modulus and Phase Angle Predictive Models for Superpave Mixtures of New Mexico
A S M Rahman, Texas A&M UniversityShow Abstract
Rafi Tarefder, University of New Mexico
The newly developed mechanistic-empirical pavement design method uses the dynamic modulus as one of the input parameters for the asphalt pavement to be designed or analyzed. In this study, new regression-based predictive models are developed to estimate dynamic modulus and phase angle of asphalt concrete from the viscosity of the asphalt binder used in the mixture. Parameters related to the aggregate gradation, and the mixture volumetric are also incorporated in these models. A total of 54 asphalt concrete mixtures with various asphalt binder grades and gradations were collected from different mixing plants and paving sites at various districts of New Mexico. The collected mixtures were then compacted, cored and sawed to cylindrical specimens for dynamic modulus and phase angle testing. The time-temperature superposition principle was applied to develop mastercurves at 70 °F (21.1 °C) reference temperature. The mastercurves were then smoothened and the parameters of these functions were then correlated to the physical attributes of the asphalt concrete samples. Finally, new predictive models are developed to estimate the dynamic modulus and phase angle of the AC mixtures typically produced in New Mexico. Statistical evaluation showed that a fairly accurate estimation of dynamic modulus and phase angle of the local mixes is possible by using these new predictive models.
Effect of Pore Water Pressure on Response of Asphalt Concrete
Maryam Shakiba, Virginia Polytechnic Institute and State UniversityShow Abstract
Masoud Darabi, University of Kansas
Dallas Little, Texas A&M University, College Station
Rapid traffic loading induces pore water pressure inside partially or fully saturated interconnected cracks and voids of asphalt concrete. The induced pore pressure contributes "extra" stresses within the pavement and accelerates crack evolution and propagation. Crack propagation facilitates dif- fusion of moisture through the solid phase and accelerates the degradation of the time-dependent stiffness and strength of asphalt concrete. Therefore, it is imperative to consider the coupled effects of pore water pressure, moisture diffusion, and mechanical loading on asphalt concrete pavement. The effect of pore water pressure is considered using the effective stress concept inside deformable media. The Biot’s approach is used and coupled to the non-linear viscoelastic and viscodamage (moisture and mechanical) constitutive relationships for asphalt concrete. The models are implemented in a finite element code PANDA (Pavement Analysis using Non-linear Damage Approach), developed at Texas A&M University. Capabilities of the proposed framework and constitutive relationships are demonstrated through simulating several realistic microstructural representations of asphalt concrete. The results of numerical simulations demonstrate how the effect of pore water pressure can intensify damage within asphalt concrete and reduce its strength. Therefore, this out- come emphasizes the importance of incorporating the effect of pore water pressure in investigating the response of asphalt concrete.
Laboratory Evaluation of a Plant-Produced High-Polymer Content Asphalt Mixture
Benjamin Bowers, Auburn UniversityShow Abstract
Stacey Diefenderfer, Virginia Transportation Research Council
Brian Diefenderfer, Virginia Transportation Research Council
Reflective cracking in asphalt overlays placed over jointed concrete pavements is of major concern in Virginia, as well as nationally, and has generated interest in various reflective crack mitigation techniques that are easy to implement. One technique is to use binder modifiers such as asphalt rubber, polymer-modified asphalt binders, or high-polymer-content modified binders. In the summer of 2014, the Virginia Department of Transportation placed a high-polymer-content (HP) asphalt mixture that was produced using an asphalt binder that contained approximately 7.5% styrene-butadiene-styrene polymer in a trial section within a subdivision as a low-risk means to assess constructability and laboratory performance. The HP mixture was evaluated in comparison to a typical 9.5mm nominal maximum aggregate size, surface asphalt mixture as a control.
Testing was performed on specimens fabricated from reheated control and HP mixture samples, as well as on specimens fabricated from site-compacted samples and road cores of the HP mixture. In addition, binder grading was performed on the control binder and modified binder. The HP binder was significantly more elastic than the control binder. Comparable dynamic moduli were found for reheated mixture specimens, although site-compacted and road core specimens from the HP mixture had lower stiffness than the control mixture. The HP mixture performed better in rutting and in fatigue. The Texas overlay test indicated similar crack resistance between the two mixtures; however, HP specimens’ measured loads were nearly half those of the control specimens.
The results of laboratory testing indicated that the mixture incorporating the HP binder should have a far greater fatigue life and rut resistance than the control mixture.
Numerical Modeling and Artificial Neural Network for Predicting J-Integral of Top-Down Cracking in Asphalt Concrete Pavement
Meng Ling, Texas A&M UniversityShow Abstract
Xue Luo, Zhejiang University
Sheng Hu, Texas A&M Transportation Institute
Fan Gu, Auburn University
Robert Lytton, Texas A&M University, College Station
Top-down cracking (TDC) is recognized as one of the major distress modes in asphalt concrete pavements. This study aims at determining the fracture parameter J-integral of the TDC, which is a critical input to predict the crack growth rate and fatigue life of pavements for this type of distress. Previous research studies demonstrated that the TDC is affected by various factors, including the complex state of high tensile or shear stresses induced by the loading at the edge of or within the tire, material properties such as the modulus gradient in the asphalt layer, moduli of the base and subgrade layers, and pavement structures. In this study, the finite element modeling (FEM) is adopted to simulate the propagation of TDC considering different combinations of these essential factors and calculate the J-integral for 194,400 cases. It is shown that the modulus gradient plays an important role in determining the J-integral, and the J-integral is not uniformly distributed within the pavement depth. Based on the database generated from the FEM, six backpropagation Artificial Neural Network (ANN) models including one input layer, two hidden layers and one output layer are developed using the same input variables and output variable as those for the FEM. The R2 value for each ANN model is more than 0.99, which indicates the goodness of fit. After determining the parameters of each ANN model, the J-integral can be predicted for any combinations of the design parameters without reconstructing the FEM.
Evaluation of Flexural Fatigue Behavior of Two Layered Asphalt Beams with Geosynthetic Interlayers Using Digital Image Correlation
Vinay Kumar V, Indian Institute of Technology, HyderabadShow Abstract
Sireesh Saride, Indian Institute of Technology, Hyderabad
In this study, flexural fatigue performance of two layered asphalt beams with and without geosynthetic interlayers has been evaluated using digital image correlation (DIC) technique. To simulate the field scenario, an old destressed pavement was considered as a bottom layer and a bituminous mix was compacted as an overlay with an appropriate tack coat and geosynthetic interlayer at the interface.
The digital images were recorded at a specific interval of load cycles during repeated load four point bending test. The displacement fields obtained from the digital images were analyzed to obtain the crack width, crack height and tensile strains and to study the crack initiation and propagation phenomena. The deformation data obtained from the DIC analysis was validated with the vertical deformations measured through LVDT. The DIC results correlate very well with the measured data. The DIC data depicted that the tensile strains were as high as 4.75% at the crack tip in the control specimen against 1.42% in a polyester grid interlayered specimen at the failure of corresponding specimens. With the inclusion of interlayers, the fatigue performance of two layered asphalt beam specimens are improved by about 39, 12 and 1.7 times for I1, I2 and I3 specimens respectively.