Neural Network-based Intelligent Compaction Analyzer for Estimating Compaction Quality of Hot Asphalt Mixes

The development and validation of a tool that can estimate the level of compaction of a Hot Mix Asphalt (HMA) pavement during its construction is addressed in this paper. Densification of asphalt pavements during their construction is usually accomplished through the use of vibratory compactors. During compaction, the compactor and the asphalt mat form a coupled system whose dynamics are influenced by the changing stiffness of the mat. In this paper, it is shown that the measured vibrations of the compactor along with the process parameters such as lift thickness, mix type, mix temperature, and compaction pressure can be used to predict the density of the asphalt mat. Contrary to existing techniques in the literature where a model is developed to fit the experimental data and to predict the density of the mat, a novel neural network based approach is adopted that is model-free and uses pattern-recognition techniques to estimate the density. During compaction of a HMA mat, the neural network then classifies the observed vibrations as those corresponding to a known level of compaction. The results also show that the analyzer can estimate the density continuously, and in realtime with accuracy levels adequate for quality control in the field. Using this tool, for the first time, the overall quality of construction of a HMA pavement can be verified thereby creating the potential to improve the quality of the roads.