Shuai Yu
Published: 2024
Total Pages: 0
Get eBook
Effective compaction is crucial for the performance and durability of asphalt pavement. Traditional field compaction, relying heavily on engineers' experience and test strips, sometimes could be problematic to achieve a unified pavement with a desirable density, especially with new materials. To address these challenges, Intelligent Compaction (IC) has been developed to equip the vibratory rollers with GPS, accelerometers, onboard computers, and infrared thermometers to facilitate the quality control of pavement compaction. This technology allows for real-time monitoring and visualization of pavement responses and temperatures, significantly improving compaction uniformity. However, accurately predicting pavement density remains challenging due to the multilayered pavement structure and the complex interactions between the roller drum and the viscoelastic asphalt mixture. To understand the compaction mechanism and improve the compaction quality of the asphalt pavement, a Microelectromechanical System (MEMS) sensor, SmartKli was employed to study the asphalt mixture compaction at the mesoscale. It was found that the compaction characteristics at the macroscale are closely related to the behavior of coarse aggregates at the mesoscale level. The particle rotation plays a critical role in the densification of the asphalt specimens. Utilizing the Discrete Element Model (DEM), the impact of mix design and particle property on kinematic behaviors was examined. The mixture gradation and particle size also greatly affect the aggregates' behavior during compaction. Based on the developed compaction mechanism, a new method for evaluating asphalt mixture workability was proposed, incorporating workability parameters, compaction curves, and statistical analysis of compaction data. By verifying with different asphalt types including Hot Mix Asphalt (HMA), Warm Mix Asphalt (WMA), and Recycled Plastic Modified Asphalt (RPMA), this method could effectively assess the influence of various factors like asphalt content, compaction temperature, and additives on mixture workability, aiding in optimizing mix design and construction conditions. Moreover, an innovative compaction monitoring system was developed to accurately predict the compaction conditions of the asphalt pavement. This system uses a wireless particle size sensor for data acquisition and a machine learning model for density prediction. Linking laboratory gyratory and field roller compaction data through particle kinematic behaviors, the system achieved high precision in density prediction with a prediction error of less than 0.7%. The results demonstrate that integrating AI and sensing data is effective for predicting asphalt mixture compaction. This system could significantly enhance the compaction quality of asphalt pavement and contribute to the comprehensive quality control and assurance of pavement construction.