CONSTRUCTING GENERATORS OF VALUES OF AN INPUT MATERIAL FLOW OF CONVEYOR-TYPE TRANSPORT SYSTEMS

Object of research is stochastic stationary input flow of material of a conveyor-type transport system. Subject of research is development of a method for generating a training data set for a neural network in a transport conveyor model. Goal of the research consists in the development of a random value generator for constructing an implementation of the input flow of material of a transport conveyor, which has specified statistical characteristics, calculated on the basis of the results of previously performed experimental measurements. The results obtained. The article considers a class of stochastic material flows of a transport system, for which it is possible to approximate the experimental implementation of a random process by an implementation corresponding to the simplest flow for the model of a random telegraph wave. A method is presented for generating values of the input material flow in order to form a data set for training a neural network in a branched transport conveyor model. The basic principles for constructing a generator of input material flow values are defined. The use of a single experimental implementation of a stochastic input material flow to calculate statistical characteristics for constructing a generator of input material flow values is justified. Dimensionless parameters have been introduced to simplify the description of stochastic input material flows and to determine similarity criteria for stochastic input material flows. The implementation of a stochastic input flow of material is presented in the form of a series expansion in coordinate functions. The law of distribution of the length of the time interval between two events of the simplest flow of events, used to approximate the implementation of the input material flow, is determined. A comparative analysis of correlation functions for experimental, approximated and generated implementations of the input material flow was carried out. The length of the time interval required to carry out experimental changes in the input material flow is justified. Estimates of the statistical characteristics of the implementations of the input material flow are given.