Machine Learning for Identifying Characteristics of Isolated, Clustered, and Pulsed Vapor Bubbles on a Heated Surface under Non-Stationary Boiling Conditions

Standard

Machine Learning for Identifying Characteristics of Isolated, Clustered, and Pulsed Vapor Bubbles on a Heated Surface under Non-Stationary Boiling Conditions. / Khan, P. V.; Levin, A. A.; Chupin, I. I. et al.

In: Energy Systems Research, Vol. 8, No. 4, 6, 29.12.2025, p. 54-64.

Research output: Contribution to journal › Article › peer-review

BibTeX

@article{87d561eea64f4207a53acb212f550752,

title = "Machine Learning for Identifying Characteristics of Isolated, Clustered, and Pulsed Vapor Bubbles on a Heated Surface under Non-Stationary Boiling Conditions",

abstract = "This paper presents an automated system for analyzing high-speed video of non-stationary nucleate boiling on an opaque steel surface. The method leverages the DenoiSeg deep learning network for robust bubble segmentation under challenging conditions (reflected light, optical distortions) and introduces an algorithm for tracking bubbles and calculating time-dependent characteristics. The system identifies and classifies bubbles into three types (isolated, clustered, and pulsating) to extract the essential boiling parameters, including nucleation site density, surface area fraction, maximum diameter, and nucleation frequency. Validation against manual key frame analysis confirms the system's accuracy. The results not only verify the significant prevalence of clustered and pulsating bubbles but also, thanks to extensive data processing, reveal trends hidden by stochastic noise, such as the growth of the maximum diameter of clusters with increasing surface temperature. The developed tool provides a reliable foundation for building predictive heat transfer models for non-stationary boiling regimes.",

keywords = "Nucleate boiling, image segmentation, machine learning, maximum bubble diameter, nucleation frequency, nucleation site density, time-averaging",

author = "Khan, {P. V.} and Levin, {A. A.} and Chupin, {I. I.} and Safarov, {A. S.}",

note = "The research was funded by the Russian Science Foundation, Grant No. 22-19-00092-П.",

year = "2025",

month = dec,

day = "29",

doi = "10.25729/esr.2025.04.0006",

language = "English",

volume = "8",

pages = "54--64",

journal = "Energy Systems Research",

issn = "2618-9992",

publisher = "Melentiev Energy Systems Institute of Siberian Branch of the Russian Academy of Sciences",

number = "4",

}

RIS

TY - JOUR

T1 - Machine Learning for Identifying Characteristics of Isolated, Clustered, and Pulsed Vapor Bubbles on a Heated Surface under Non-Stationary Boiling Conditions

AU - Khan, P. V.

AU - Levin, A. A.

AU - Chupin, I. I.

AU - Safarov, A. S.

N1 - The research was funded by the Russian Science Foundation, Grant No. 22-19-00092-П.

PY - 2025/12/29

Y1 - 2025/12/29

N2 - This paper presents an automated system for analyzing high-speed video of non-stationary nucleate boiling on an opaque steel surface. The method leverages the DenoiSeg deep learning network for robust bubble segmentation under challenging conditions (reflected light, optical distortions) and introduces an algorithm for tracking bubbles and calculating time-dependent characteristics. The system identifies and classifies bubbles into three types (isolated, clustered, and pulsating) to extract the essential boiling parameters, including nucleation site density, surface area fraction, maximum diameter, and nucleation frequency. Validation against manual key frame analysis confirms the system's accuracy. The results not only verify the significant prevalence of clustered and pulsating bubbles but also, thanks to extensive data processing, reveal trends hidden by stochastic noise, such as the growth of the maximum diameter of clusters with increasing surface temperature. The developed tool provides a reliable foundation for building predictive heat transfer models for non-stationary boiling regimes.

AB - This paper presents an automated system for analyzing high-speed video of non-stationary nucleate boiling on an opaque steel surface. The method leverages the DenoiSeg deep learning network for robust bubble segmentation under challenging conditions (reflected light, optical distortions) and introduces an algorithm for tracking bubbles and calculating time-dependent characteristics. The system identifies and classifies bubbles into three types (isolated, clustered, and pulsating) to extract the essential boiling parameters, including nucleation site density, surface area fraction, maximum diameter, and nucleation frequency. Validation against manual key frame analysis confirms the system's accuracy. The results not only verify the significant prevalence of clustered and pulsating bubbles but also, thanks to extensive data processing, reveal trends hidden by stochastic noise, such as the growth of the maximum diameter of clusters with increasing surface temperature. The developed tool provides a reliable foundation for building predictive heat transfer models for non-stationary boiling regimes.

KW - Nucleate boiling

KW - image segmentation

KW - machine learning

KW - maximum bubble diameter

KW - nucleation frequency

KW - nucleation site density

KW - time-averaging

UR - https://www.scopus.com/pages/publications/105029555098

UR - https://www.mendeley.com/catalogue/d7ee7714-6d64-3138-9461-59ff8a2bd021/

U2 - 10.25729/esr.2025.04.0006

DO - 10.25729/esr.2025.04.0006

M3 - Article

VL - 8

SP - 54

EP - 64

JO - Energy Systems Research

JF - Energy Systems Research

SN - 2618-9992

IS - 4

M1 - 6

ER -

ID: 74603521