A flow pattern recognition method for gas-liquid two-phase flow based on dilated convolutional channel attention mechanism

Tallennettuna:
Bibliografiset tiedot
Julkaisussa:PLoS One vol. 20, no. 6 (Jun 2025), p. e0325784
Päätekijä: Liu, Jie
Muut tekijät: Wu, Yang
Julkaistu:
Public Library of Science
Aiheet:
Linkit:Citation/Abstract
Full Text
Full Text - PDF
Tagit: Lisää tagi
Ei tageja, Lisää ensimmäinen tagi!

MARC

LEADER 00000nab a2200000uu 4500
001 3223867926
003 UK-CbPIL
022 |a 1932-6203 
024 7 |a 10.1371/journal.pone.0325784  |2 doi 
035 |a 3223867926 
045 2 |b d20250601  |b d20250630 
084 |a 174835  |2 nlm 
100 1 |a Liu, Jie 
245 1 |a A flow pattern recognition method for gas-liquid two-phase flow based on dilated convolutional channel attention mechanism 
260 |b Public Library of Science  |c Jun 2025 
513 |a Journal Article 
520 3 |a Addressing the issue of insufficient key feature extraction leading to low recognition rates in existing deep learning-based flow pattern identification methods, this paper proposes a novel flow pattern image recognition model, Enhanced DenseNet with transfer learning (ED-DenseNet). The model enhances the deep feature extraction capability by introducing a multi-branch structure, incorporating an ECA attention mechanism into Dense Blocks and dilated convolutions into Transition Layers to achieve multi-scale feature extraction and refined channel information processing. Considering the limited scale of the experimental dataset, pretrained DenseNet121 weights on ImageNet were transferred to ED-DenseNet using transfer learning. On a gas-liquid two-phase flow image dataset containing Annular, Bubbly, Churn, Dispersed, and Slug flow patterns, ED-DenseNet achieved an overall recognition accuracy of 97.82%, outperforming state-of-the-art models such as Flow-Hilbert–CNN, especially in complex and transitional flow scenarios. Additionally, the model’s generalization and robustness were further validated on a nitrogen condensation two-phase flow dataset, demonstrating superior adaptability compared to other methods. 
653 |a Flow distribution 
653 |a Feature extraction 
653 |a Pattern recognition 
653 |a Flow pattern 
653 |a Datasets 
653 |a Data processing 
653 |a Nuclear reactors 
653 |a Identification methods 
653 |a Neural networks 
653 |a Signal processing 
653 |a Two phase flow 
653 |a Information processing 
653 |a Slug flow 
653 |a Deep learning 
653 |a Machine learning 
653 |a Multiphase flow 
653 |a Transfer learning 
653 |a Transition layers 
653 |a Environmental 
700 1 |a Wu, Yang 
773 0 |t PLoS One  |g vol. 20, no. 6 (Jun 2025), p. e0325784 
786 0 |d ProQuest  |t Health & Medical Collection 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3223867926/abstract/embedded/75I98GEZK8WCJMPQ?source=fedsrch 
856 4 0 |3 Full Text  |u https://www.proquest.com/docview/3223867926/fulltext/embedded/75I98GEZK8WCJMPQ?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3223867926/fulltextPDF/embedded/75I98GEZK8WCJMPQ?source=fedsrch