Open AccessReal-time neural network system for non-destructive control of asphalt mixtures compaction
Author(s) -
Zh.I. Nabizhanov,
Андрей Петрович Прокопьев,
Vladimir I. Ivanchura,
R. T. Emelyanov
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1181/1/012021
Subject(s) - artificial neural network , compaction , backpropagation , normalization (sociology) , real time control system , control system , asphalt , computer science , database normalization , levenberg–marquardt algorithm , engineering , artificial intelligence , control (management) , geotechnical engineering , pattern recognition (psychology) , materials science , electrical engineering , sociology , anthropology , composite material
The problem is considered of designing a neural network system for control of road materials compaction in real time. The system is designed to ensure the functioning of non-destructive technology in road construction. A model of an artificial neural network (ANN) for monitoring the quality of compaction of road materials during the operation of vibrating rollers in real time is obtained. The system provides normalization of the input data of the ANN model. For the design of the ANN, the following methods were used: the least mean square deviation method; Levenberg-Marquardt algorithm; backpropagation. The structure of a non-destructive neural network system for monitoring compaction in real time is obtained. The results of numerical simulation of a system with an ANN for predicting the compaction coefficient of road materials are presented