MODELS AND PROCEDURES FOR CLASSIFICATION AND FORECASTING OF PROCESS PARAMETERS ACCORDING TO CHAOTIC DYNAMICS

Abstract

The report investigates the processes of modeling and short-term prediction of nondeterministic time sequences, which are anti-persistent time series (AРТS) according to the classification based on the Hearst parameter. The subject of analysis is the numerical procedures for classification and forecasting the parameters of such models. The aim of the work was to increase the efficiency and accuracy of methods and algorithms for classification, modeling and forecasting of AРТS. The objectives of the study were to develop by aggregating the levels of AРТS mathematical models for the classification of time series, as well as the formation of algorithmic and software tools. The report also presents a tool software environment that provides a correct study of modeling algorithms and classification of AРТS.