Сибирский федеральный университет

Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций

Конференция: Springer Science and Business Media Deutschland GmbH; 14 October 2020 through 17 October 2020; 14 October 2020 through 17 October 2020

Год издания: 2020

Идентификатор DOI: 10.1007/978-3-030-63322-6_49

Ключевые слова: artificial neural networks, classification of failures, data analysis, technical diagnostics

Аннотация: The article considers the problem of choosing a data analysis technology for designing a system for identification and predicting failures of technological equipment based on vibration monitoring data. The task of analyzing vibration monitoring data is solved in relation to the technological equipment of a fuel-oriented oil refinery. The article presents the results of the sensitivity analysis of the models for determining the type and failures with respect to various vibration parameters recorded by a system of vibration sensors. The results of the analysis based on data on failures show a difference in determining the most significant factors for different methods of data analysis. In the article for designing models for determining failure types, methods of discriminant analysis, decision trees, multidimensional regression splines, and a neural network approach are considered. As a result of applying the methods to the data set on failures of technological pumping equipment, it was determined that the method based on artificial neural networks is the most effective. Taking into account the use of tools for the automatic construction of neural network classifiers, such models can be further used in an automatically deployed global system for ensuring the reliability of technological equipment in oil and gas production. #COMESYSO1120 © 2020, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.

Журнал: Advances in Intelligent Systems and Computing

Выпуск журнала: Vol. 1294

Номера страниц: 598-605

ISSN журнала: 00253159

• Bukhtoyarov V. (Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, Russian Federation; Siberian Federal University, Krasnoyarsk, Russian Federation)
• Zyryanov D. (Siberian Federal University, Krasnoyarsk, Russian Federation)
• Tynchenko V. (Siberian Federal University, Krasnoyarsk, Russian Federation)
• Bashmur K. (Siberian Federal University, Krasnoyarsk, Russian Federation)
• Petrovsky E. (Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, Russian Federation)

• Scopus
• Ядро РИНЦ (eLIBRARY.RU)