Novosibirsk State Technical University (NSTU) has created a digital model of a complex technological installation to improve the quality of management and optimize refining processes.
The distillation column is a key device in the process of separating crude oil into fractions, its operation is based on the difference in boiling points of the components. The feedstock is fed into the column for subsequent separation. When its lower part is reached, thermal evaporation of the components occurs, followed by fractional condensation on contact devices (plates) located in the appropriate temperature zones. As part of the technological process, the target fractions are isolated: gasoline, kerosene, naphtha, gas oil and fuel oil.
To ensure flexible process control and accurate selection of target fractions (gasoline, naphtha, kerosene), the system is equipped with shut-off valves, in particular valves. They are responsible for controlling the flow of steam and phlegm (liquid), ensuring stable operation of the separation process.
Anastasia Stoyanova, a graduate student at the Faculty of Mechatronics and Automation at NSTU-NETI, notes that the output volume of a certain fraction can be increased by 5-7% by adjusting the degree of valve opening. Some of the raw materials will be retained in the column and redistributed to neighboring outlets. An increase in the output of specific fractions may be necessary to generate economic profit based on the current market demand for a particular product and its cost.
A master's student from NSTU-NETI has developed a mathematical model that provides the physics of the rectification process, when raw materials are fed to the column and divided into fractions. In the created digital twin, it is possible to virtually adjust the position of the valves and evaluate the effect on the fraction output, which allows you to work out control scenarios without risking real equipment. The output of the target fractions in the distillation column is regulated by the degree of valve opening through control of the phlegm number and flow distribution in the column. In addition, an automated workplace was created in the SCADA system to monitor the system's performance and control the technological process.
"Now I can set the degree of valve opening arbitrarily and get certain values of the fraction output. The task that will be solved in the future is for the algorithm to select the degree of valve opening itself, that is, to calculate and produce the result: at what values of the degree of valve opening there will be the greatest yield of a certain fraction," Anastasia Stoyanova said.
It is planned to finalize the digital model by mid-2027.