Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 144Heat Transfer and Boiling StudiesFlow Measurement and AnalysisOffshore Engineering and Technologies

The DOI matches its Crossref record; 47 of the 47 references it cites were found in a registry or an index.

Prediction of two-phase flow patterns in upward inclined pipes via deep learning

Open access
Lin, Zi, Liu, Xiaolei, Lao, Liyun, Liu, Hengxu·2020·Energy
DOI 10.1016/j.energy.2020.118541
Open this paper in CoChat — free accountOpen PDF

Abstract

The industrial process involving gas liquid flows is one of the most frequently encountered phenomena in the energy sectors. However, traditional methods are practically unable to reliably identify flow patterns if additional independent variables/parameters are to be considered rather than gas and liquid superficial velocities. In this paper, we reported an approach to predict flow pattern along upward inclined pipes (0–90°) via deep learning neural networks, using accessible parameters as inputs, namely, superficial velocities of individual phase and inclination angles. The developed approach is equipped with deep learning neural network for flow pattern identification by experimental datasets that were reported in the literature. The predictive model was further validated by comparing its performance with well-established flow regime forecasting methods based on conventional flow regime maps. Besides, the intensity of key features in flow pattern prediction was identified by the deep learning algorithm, which is difficult to be captured by commonly used correlation approaches.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1016/j.energy.2020.118541

Cites

47 of 47 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Scatter matrix of selected features in the experimental datasets, including inclination angles, superficial velocities of liquid, superficial velocities of air, and corresponding flow patterns; note that, the observed flow pattern categories were digitalized as 1 - dispersed bubble, 2 - stratified smooth, 3 - stratified wavy, 4 - annular, 5 - intermittent, and 6 - bubble in the subplot of (d).
Fig. 1 Scatter matrix of selected features in the experimental datasets, including inclination angles, superficial velocities of liquid, superficial velocities of air, and corresponding flow patterns; note that, the observed flow pattern categories were digitalized as 1 - dispersed bubble, 2 - stratified smooth, 3 - stratified wavy, 4 - annular, 5 - intermittent, and 6 - bubble in the subplot of (d).CC-BY-SA · From the PDF
Fig. 2 Heat map of features in the experimental datasets; note that, V(gas) is representing the superficial velocity of air and V(liquid) is representing the superficial velocity of water.
Fig. 2 Heat map of features in the experimental datasets; note that, V(gas) is representing the superficial velocity of air and V(liquid) is representing the superficial velocity of water.CC-BY-SA · From the PDF
Fig. 3 Visualization of the deep learning configuration.
Fig. 3 Visualization of the deep learning configuration.CC-BY-SA · From the PDF
Fig. 4 Variations of training and testing MSEs along 4000 epochs in the designed neural networks.
Fig. 4 Variations of training and testing MSEs along 4000 epochs in the designed neural networks.CC-BY-SA · From the PDF
Fig. 5 Comparison of flow pattern predictions between predictive model (dotted points) and experimental observations (full lines).
Fig. 5 Comparison of flow pattern predictions between predictive model (dotted points) and experimental observations (full lines).CC-BY-SA · From the PDF
Fig. 6 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in horizontal (0°) pipe.
Fig. 6 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in horizontal (0°) pipe.CC-BY-SA · From the PDF
Fig. 7 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in upward inclined (1°) pipe.
Fig. 7 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in upward inclined (1°) pipe.CC-BY-SA · From the PDF
Fig. 8 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in upward inclined (30°) pipe.
Fig. 8 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in upward inclined (30°) pipe.CC-BY-SA · From the PDF
Fig. 9 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in upward inclined (80°) pipe.
Fig. 9 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in upward inclined (80°) pipe.CC-BY-SA · From the PDF
Fig. 10 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in vertical (90°) pipe.
Fig. 10 Comparisons of flow regime maps generated by deep learning model (dotted points) and unified model (full lines) for two-phase flow in vertical (90°) pipe.CC-BY-SA · From the PDF
Image 11
CC-BY-SA · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 47 references

Cited by

Related

See all

Metadata

Authors
Lin, Z., Liu, X., Lao, L., & Liu, H.
Affiliations
  • Northumbria University (GB)
  • University of Glasgow (GB)
  • Cranfield University (GB)
  • Harbin Engineering University (CN)
Corresponding author
Liu, Xiaolei
ORCID
Venue
Energy
Publisher
Elsevier BV
Published
2020-11
Volume
210
Pages
118541
ISSN
0360-5442 (ISSN-L)
DOI
10.1016/j.energy.2020.118541
OpenAlex
W3049349475
Cited by
142
References
47
Citations by year
2026: 21 · 2025: 24 · 2024: 30 · 2023: 22 · 2022: 23 · 2021: 17 · 2020: 5
FWCI
8.05
Type
journal article
Language
en
Open access
Licence
  • cc-by-sa
  • www.elsevier.com/tdm/userlicense/1.0/ (tdm)
  • www.elsevier.com/legal/tdmrep-license (tdm)
Locations
  • Energy (publishedVersion)
  • ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam) (acceptedVersion, cc-by-sa)
  • Northumbria Research Link (Northumbria University) (submittedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • CERES (Cranfield University) (submittedVersion, cc-by-nc-nd)
Topics
  • Physical Sciences › Engineering › Mechanical Engineering › Heat Transfer and Boiling Studies
  • Physical Sciences › Physics and Astronomy › Instrumentation › Flow Measurement and Analysis
  • Physical Sciences › Engineering › Ocean Engineering › Offshore Engineering and Technologies
Keywords
two-phase flow, flow regime prediction, gas-liquid two-phase flow, superficial gas velocity, inclination angle, artificial neural networks, flow regime map, feature importance

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.