Neural network control system for compaction of road materials of pavers
Abstract
Neural network control system for compaction of road materials of pavers
Incoming article date: 19.09.2021The problem of creating a neural network system of automatic control (ACS) of the compaction process for asphalt pavers (AP) is considered. This task is aimed at improving the performance of the AP and the quality of road surfaces of highways. The model of inverse neurocontrol is implemented. When training an artificial neural network (ANN), an exit error of the control object was used. Input information signals of the neural network ACS: the speed of the paver; type of asphalt concrete mixture; layer thickness; force in the tamper pusher; acceleration of the vibrating plate. The is presented results of a computational experiment in the MATLAB/Simulink program, which showed good convergence with experimental data from field tests of pavers.
Keywords: automatic compaction control, non-destructive technologies, artificial neural networks, road construction, pavers