Email this article Combustion Regime Monitoring by Flame Imaging and Machine Learning
- Authors: Abdurakipov S.S.1,2, Gobyzov O.A.1,2, Tokarev M.P.1,2, Dulin V.M.1,2
-
Affiliations:
- Kutateladze Institute of Thermophysics, Siberian Branch
- Novosibirsk State University
- Issue: Vol 54, No 5 (2018)
- Pages: 513-519
- Section: Modeling in Physical and Technical Research
- URL: https://journals.rcsi.science/8756-6990/article/view/212588
- DOI: https://doi.org/10.3103/S875669901805014X
- ID: 212588
Cite item
Open Access
Access granted
Subscription Access
Abstract
A method for automatic determination of combustion regimes using flame images on the basis of a convolutional neural network on labeled data is under consideration. It is shown that the accuracy of regime classification reaches 98% on the flame images of a gas burner. The results of the operation of the convolutional neural network and classification using different linear models are compared.
About the authors
S. S. Abdurakipov
Kutateladze Institute of Thermophysics, Siberian Branch; Novosibirsk State University
Email: oleg.a.g.post@gmail.com
Russian Federation, pr. Akademika Lavrent’eva 1, Novosibirsk, 630090; ul. Pirogova 2, Novosibirsk, 630090
O. A. Gobyzov
Kutateladze Institute of Thermophysics, Siberian Branch; Novosibirsk State University
Author for correspondence.
Email: oleg.a.g.post@gmail.com
Russian Federation, pr. Akademika Lavrent’eva 1, Novosibirsk, 630090; ul. Pirogova 2, Novosibirsk, 630090
M. P. Tokarev
Kutateladze Institute of Thermophysics, Siberian Branch; Novosibirsk State University
Email: oleg.a.g.post@gmail.com
Russian Federation, pr. Akademika Lavrent’eva 1, Novosibirsk, 630090; ul. Pirogova 2, Novosibirsk, 630090
V. M. Dulin
Kutateladze Institute of Thermophysics, Siberian Branch; Novosibirsk State University
Email: oleg.a.g.post@gmail.com
Russian Federation, pr. Akademika Lavrent’eva 1, Novosibirsk, 630090; ul. Pirogova 2, Novosibirsk, 630090
Supplementary files
