Modeling of Oil Bitumen Quality Parameters Using Machine Learning Algorithms.

The paper considers approaches, principles, and results of modeling the quality parameters of petroleum bitumen using machine learning algorithms based on recurrent neural networks. It is shown that machine learning algorithms can be effectively used in practice for oil refining processes. Various problems involved in data processing, as well as selection of variables and suitable neural network architecture for solving a particular problem, are considered. Further research directions are outlined. [ABSTRACT FROM AUTHOR]