Clustering problem in modeling the composition of a mixture of catalytic reforming of gasoline

Return to list of articles
Doklady Bashkirskogo Universiteta. 2021. Volume 6. No. 5. pp. 302-306.

Authors

Alekseeva V. A.*
Ufa State Petroleum Technological University
1 Kosmonavtov Street, 450064 Ufa, Republic of Bashkortostan, Russia
Koledina K. F.
Ufa State Petroleum Technological University; Institute of Petrochemistry and Catalysis of the UFIC RAS
1 Kosmonavtov Street, 450064 Ufa, Republic of Bashkortostan, Russia; 141 Prospekt Oktyabrya, 450075 Ufa, Republic of Bashkortostan, Russia

Abstract

The article is devoted to the study of the analysis of octane numbers of hydrocarbons of the process of catalytic reforming of gasoline by clustering methods in the Python system. It is known that a large number of individual hydrocarbons (up to 300 different components) are present in catalytic reforming. Which significantly complicates the simulation. Therefore, they are grouped together. For groups, it is necessary to determine the characteristics - octane numbers in the first place. To do this, it is necessary to group hydrocarbons according to the values of octane numbers. To do this, we will use the clustering task. The task of splitting a given sample of objects into disjoint subsets, called clusters, so that each cluster consists of similar objects, and the objects of different clusters differ significantly. To solve this problem, we will use the clustering problem with two machine learning algorithms, this is the hierarchical clustering algorithm and the K-Means algorithm.

Keywords

  • catalytic reforming
  • clustering algorithms
  • hierarchical
Return to list of articles