Journal "Software Engineering"
a journal on theoretical and applied science and technology
ISSN 2220-3397
Issue N8 2021 year
The results of the research on the creation of an automatic compaction control system (ACCS) for pavers in real time are considered. The research is based on the methods of artificial neural networks (ANN). In this paper, an ANN model is obtained, with the help of which it is possible to determine the compaction coefficient (CC) of an asphalt mixture. The input variables of the ACCS are the velocity of movement of the paver, the frequency of impacts of the tamper, the force in the pusher of the tamper, the type of mixture, the thickness of the layer. The results of a computational experiment on the calculation of Cc in real time are presented. The ANN is able to explain more than 98 % of the measured data.