On the choice of the method of dynamic rationing of energy resources in oil refineries

Abstract

The article discusses the possibility of calculating the expected energy demand based on big data and machine learning for the energy technological processes in oil refineries. In order to obtain predictive data, linear regression, machine learning, and neural networks are proposed to be used to build a mathematical model. The advantages and disadvantages of these methods are discussed, and the accuracy of the models is compared with the possibility of interpreting them. Thanks to the use of advanced statistical methods, the variability of energy consumption can be interpreted through factor analysis. Through pilot tests, the practical significance of these proposed methods for their practical use in an energy management system is demonstrated, as well as the transition to statistical control of the process.

Keywords

• rationing of energy resources
• fuel and energy resources
• rationing methods
• linear regression
• machine learning
• deep learning
• factor analysis
• energy management system