Application of neuro-fuzzy systems technology for control the density of asphalt concrete mixtures in the paving process

Abstract

The increase in the number of cars and trucks, and the intensity of their use, directly affects the asphalt-concrete road surfaces of roads. This is manifested in the reduction of overhaul terms of operation and large financial costs. Improving the quality and increasing the service life of asphalt concrete pavements of roads is a national economic problem. This problem is solved by improving the regulatory framework, improving the properties of road materials, optimizing and automating technological processes in road construction. Road roller density monitoring and control systems are based on Intelligent Compaction and Continuous Compaction Control technologies. In the Russian Federation, the quality of road surface construction largely depends on the results of the work of pavers, who ensure the acceptance, laying and compaction of mixtures. Many defects in the operation of road surfaces are eliminated by high-quality compaction. Automatic control and control of the sealing process for asphalt pavers have not been developed.The aim of the study is to build a density management system for asphalt pavers based on an intelligent automatic control system with feedback. Real-time prediction of the volume density of the layer, effective manual (operator) and automatic density control is organized to achieve the required quality indicators.The results of the development of a new system for intelligent control of the density asphalt concrete mixtures by pavers is considered. The automatic control system with feedback includes a continuous density control system designed to calculate the seal quality index based on the implementation of the neural network structure algorithm in real time, as well as a neuro-fuzzy PID regulator. The study considers a system with models of control objects of high order – the fourth and sixth orders. The structure of a neuro-fuzzy network of the ANFIS type is proposed. The generation of a fuzzy output system is performed on the basis of the lattice partition method. ANFIS training is performed by a hybrid method on an array of variables obtained as a result of modeling an automatic control system with an analog PID controller. Possible modes of use an intelligent compaction control system: continuous automatic control with manual control of the operating modes of compaction; automatic density control. Automatic measure of density and control compaction modes is aimed at improving the quality of asphalt concrete pavements of roads and improving the quality of asphalt pavements efficiency of technological processes road construction.