Approach to gas consumption process forecasting on the basis of a mathematical model in the form of a random cyclic process

Abstract

In the article, the approach to gas consumption process prediction on the basis of a mathematical model in the form of a cyclic random process is considered. The prediction is based on statistical information for previous years of gas consumption. An additive combination of three components was used as a mathematical model of gas consumption process: a cyclic random process, a trend component and a stochastic residue. The first component of the mathematical model in the form of a cyclic random process takes into account the scale factors, information about which is considered at the stage of prediction. Based on the caterpillar decomposition, ten components of the singular decomposition are obtained. The sum of nine components of singular decomposition forms a cyclic component – a cyclic random process. This component takes into account the scale factors of the amplitude of gas consumption process in each segment-cycle. The trend component of the mathematical model is the second component of the singular decomposition, and the stochastic residue is formed on the basis of the difference between the values of the studied gas consumption process and the sum of the cyclic and trend components. Computer simulation of realization of cyclic component of gas consumption process is carried out in the work, and also the annual forecast of gas consumption is made. The prediction results are compared with the real gas consumption process (information for last yearʼs gas consumption was used). This paper does not take into account the effects of climatic conditions on the gas consumption process, but this is going to be done in further research, which will improve the accuracy of computer simulation and prediction.