Predicting the parameters of the output flow of conveyor systems

Abstract

In this article, the problem of constructing models of multi-section conveyor-type transport systems using a neural network is considered. The analysis of models of long-distance multi-section transport systems, which are used to design control systems of the flow parameters from the point of view of reducing the unit costs of material transportation, is presented. The areas of the models' application and associated limitations are demonstrated. The advantages of using neural networks for developing multi-section transport systems models are shown. The influence of the initial distribution of materials along the transport route and the presence of a transport delay in the system on the quality of prediction of the transport system flow parameters are estimated. The effect of the use of speed sensors located on the inner and output sections in order to improve the quality of prediction of the transport system flow parameters is analyzed. It is shown that the use of speed sensors in conveyors with belt speed control can significantly improve the quality of predict.