FORMATION OF STATION ELEMENT CHARACTERISTICS ON THE BASIS OF MACHINE LEARNING METHODS

Olimpiev Sergey Aleksandrovich, Neklyudov Alexey Vasilyevich

Purpose: Deviations of equipment operation parameters can lead to a decrease in the efficiency of individual technological elements of the plant. In this regard, the task of determining correct parameters of equipment operation remains quite important in the field of power engineering. The article is devoted to the study of the approach to the formation of the characteristic surface of the plant element using the machine learning method. The article considers the algorithm for building the characteristic surface of the sharp steam flow rate parameter for a heat recovery turbine, and also evaluates the quality of the developed model in accordance with the accuracy indicators. Methods: The linear regression method is a statistical linear regression method, which is used to construct a linear model describing the relationship between a dependent variable and a set of other independent variables. Results: Algorithm for constructing the characteristic surface of the station element, calculation of the accuracy indicators of the model. Conclusions of the study: The approach based on the method of linear regression for the construction of the characteristic surface of the station element is shown. The results of the work can be applied in studies related to the process of updating the equipment parameters of station elements.

Keywords:characteristic area, regression analysis, machine learning, turbine, thermal power plant