Fatigue damage evolution processes of asphalt pavement under different loading modes : from laboratory assessment to full-scale apt investigation

Abstract

Fatigue damage critically threatens the integrity of asphalt pavements and remains a central challenge for reliable performance prediction. This study aims to propel this field forward by developing a predictive model that not only intricately characterizes the fatigue damage process but also reveals the role of fatigue loading modes in shaping the damage progression. This novel approach not only sheds light on the mechanisms driving fatigue progression but also bridges the gap between laboratory fatigue assessments and the fatigue behaviors observed in asphalt layers of in-service pavements. Central to the findings is the realization that fatigue damage is not solely governed by tensile fatigue loading but rather by complex tensile-shear fatigue loading modes. This key insight is pivotal for pinpointing critical damage locations within pavement structures, offering a strategic advantage for enhancing pavement design and maintenance practices. To achieve the objective, the first task conducts an in-depth laboratory program using indirect tensile fatigue (ITFT) and four-point bending (4PB) tests. It initially examined the impact of various loading parameters, including tensile fatigue loading modes, on the initial stiffness and fatigue life of asphalt mixtures. Consequently, a versatile fatigue life equation was formulated across tensile fatigue loading modes and damage states. Subsequently, to gain a comprehensive understanding of fatigue damage progression in the laboratory setting, seven nonlinear models were explored, particularly highlighting the structural behavior function (SBF) and the Weibull survival function (WSF). This methodological approach led to a foundation for accurately modeling fatigue damage development, with the SBF distinguished from others for its reliability and applicability. Additionally, the analysis of various loading parameters, most notably the fatigue loading mode, was intricately woven into the predictive models, significantly enhancing the interpretability of the model parameters and providing a deeper insight into asphalt mixture fatigue behaviors. The second task of this study expanded to full-scale accelerated pavement testing (F-sAPT) on in-situ asphalt layers with various types of bases, including semi-rigid and granular bases. A key fatigue damage indicator was assessed using the falling weight deflectometer (FWD), portable seismic pavement analyzer (PSPA), and uniaxial compression tests. These measurements were further refined by applying temperature correction and frequency adjustment methodologies. The investigation then delved into analyzing the damage progressions across different depths and orientations of asphalt layers. Both the SBF and WSF proved effective in modeling the measured deterioration processes, with the SBF being recognized as accurately depicting damage evolution within pavement structures. The study's integration of field measurements with advanced analytical techniques enhanced the understanding of pavement deterioration dynamics, validating methods used and offering practical insights for future pavement damage assessment. In the third phase of the study, a meticulous analysis of the mechanical mechanisms influencing fatigue damage in asphalt layers was conducted, leveraging the findings from F-sAPT. This investigation led to a pivotal discovery: In-situ fatigue damage in asphalt layers is not exclusively governed by tensile fatigue loading but is profoundly impacted by complex tensile-shear fatigue loading modes. The identification of these modes is critical for accurately pinpointing key damage points within asphalt layers. Additionally, a novel tensile-shear loading mode factor was proposed. This enhancement provides a more comprehensive understanding of asphalt layer fatigue and its practical implications. The fourth task was dedicated to bridging the gap between laboratory findings and in-situ observations regarding asphalt fatigue damage. A key strategy involved the alignment of damage progression curves from controlled laboratory experiments with those observed under real-world conditions. This entailed a systematic comparison and correlation of model parameters derived from both laboratory tests and F-sAPT data. By substituting in-situ loading parameters into the laboratory-based model parameter interpretation equations, the study effectively achieved a calibrated lab-to-field translation. This approach not only validated the laboratory findings but also provided a robust framework for predicting in-situ pavement performance.