Xue Luo

Research on bituminous material fatigue has traditionally focused on tensile or shear damage of bitumen and asphalt mixtures, neglecting the critical bitumen–aggregate interfaces where microcracks initiate. Addressing this gap, the pull-off fatigue crack (POF-C) model was built to predict crack propagation at these interfaces under pull-off cyclic loading. The model, based on continuum damage mechanics principles, integrates force equilibrium and dissipated strain energy equilibrium. Pull-off fatigue tests were conducted on interfaces using limestone, tuff, and basalt aggregates, with #70 matrix bitumen and styrene–butadiene–styrene copolymer-modified bitumen, at temperatures of 15°C and 20°C, and with bitumen film thicknesses ranging from 0.2 mm to 0.8 mm. Dynamic modulus and phase angle data informed the model inputs. Predicted crack sizes closely matched measured results on fractured surfaces, demonstrating less than 2% prediction error. Scanning electron microscope tests confirmed the model's validity, showing numerous circular mesh depressions on fracture surfaces. The POF-C model accurately forecasts POF-C lengths across varied conditions, revealing three distinct stages of crack propagation: a rapid growth (∼0.025 mm/cycle), a stable expansion stage (<0.025 mm/cycle), and a slow fatigue stage (∼0 mm/cycle). The fatigue mechanism involves the development of microdamage into microcracks, their nucleation and aggregation, and macrocrack throughout the entire bitumen–aggregate interface.

The viscoelastic damage evolution at the microscale particle interfaces of asphalt mortar or mixtures fundamentally determines the material's fatigue crack growth and failure at the macroscale. However, existing microscale damage models are often based on empirical assumptions that depend on interparticle stresses or forces, which are inaccurate or even incorrect for viscoelastic asphalt materials. In these materials, interfacial crack growth is governed by the viscoelastic energy release rate. To address the limitations of current models, a microscale viscoelastic fatigue damage model was developed using a pseudo J-integral-based Paris’ law and implemented in a discrete element model of asphalt mortar using the PFC2D program. The viscoelastic constitutive behavior was represented by a generalized Maxwell model, and the relaxation moduli were determined through a uniaxial compressive dynamic modulus test. The Paris’ law coefficients were calibrated by comparing model predictions with experimental results from indirect tensile fatigue tests of the material. The results show that the simulated fatigue life and crack area closely match laboratory test data, with an error margin within 15%. During the simulation of microscopic IDT fatigue damage, cracks hinder the horizontal transfer of forces within the cracked region, leading to stress concentrations in surrounding particles and a marked increase in their relative displacement. The connection of upper and lower cracks significantly reduces the specimen's load-bearing capacity. The variation in the number of contact breaks with fatigue load cycles is unaffected by the type of asphalt but is influenced by the applied stress level. These findings demonstrate that the pseudo J-integral-based Paris’ law, when applied at particle interfaces, can effectively model crack growth at the microscale and accurately predict the fatigue damage performance of viscoelastic asphalt materials at the macroscale.

Fatigue cracking is one of the main pavement failures, which makes accurate fatigue life prediction for the design and maintenance of asphalt pavements crucial. The majority of traditional prediction methods are based entirely on the laboratory fatigue test, without considering the field condition and maintenance data. This paper aims to propose a hybrid approach to fill this gap. The key idea is that the damage condition is back-calculated by an artificial intelligence-based finite-element (FE) model updating using field-monitoring information (data-driven component), which is used to update the parameters in the mechanistic composition-specific fatigue life prediction equation (model-driven component). The laboratory test of field cores gives the material non-destructive properties. The simulated pavement response subjected to truck loading shows good agreement with measured values, which indicates that the verified constitutive relationship could be used in the data-driven component. Furthermore, in view that the fatigue test is time- and money-consuming, this paper proposes a non-test estimation of the fatigue characteristic curve based on FE simulation of a repeated direct tension test. Three test pavement sections were employed as case studies. Results showed that the predicted fatigue life changes with the service time. At the early age, semi-rigid pavement has a larger fatigue life than flexible and inverted pavements. This article is part of the theme issue 'Artificial intelligence in failure analysis of transportation infrastructure and materials'.

The service life and service quality of transportation infrastructures are crucial for the regular function and public safety of transportation systems. Thus, the intelligent and accurate evaluation, prediction, and analysis of transportation infrastructure service conditions have always been hot topics for transportation engineers, where such problems are sometimes hard to be solved using traditional approaches, considering different factors including complex traffic conditions, severe environmental conditions, etc. In recent years, intelligent technologies, including advanced sensors, novel artificial intelligent approaches (deep learning, etc.), the IoT-based monitoring and analysis platform, auto piloting, green energy, synergistic technologies, and systems integrating both traffic and infrastructures monitoring, have emerged as powerful tools in transportation engineering area. These efforts have greatly advanced the service convenience of public transportation, prolonged the service life of transportation infrastructures, and improved the efficiency and safety of transportation systems.