Initial study of application of Neural Networks to predict flow patterns in multiphase flow
DOI:
https://doi.org/10.14808/sci.plena.2022.084802Keywords:
multiphase flow, flow pattern, neural networkAbstract
The study of flow patterns in multiphase flow is widely used in the oil industry. Many tools are being developed to help predict these flow patterns. Artificial Intelligence is a tool with high performance and great use to characterize physical phenomena and can be used to predict these flow patterns in multiphase flow. We developed an Artificial Neural Network to recognize and predict the flow patterns of multiphase flow using an experimental database. In addition to using the neural network, data processing was also carried out to improve the tool's effectiveness. For a complete analysis, we trained the network with several combinations of the patterns, the patterns being: intermittent, annular and stratified. We did not use the bubble pattern due to the difference in the amount of data available between this and the other patterns. The net was up to 90% effective in some combinations of the analysis between flow patterns. Analyzing the 3 patterns together, the network showed an efficiency of approximately 67%. We analyzed the results of the net error behavior during its training and the hit rate after training. In this article, we used the artificial neural network as a tool to predict the flow pattern in the multiphase flow, with the maximum available flow parameters.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2022 Jair Rodrigues Neyra, Thiago Rafael da Silva Moura
![Creative Commons License](http://i.creativecommons.org/l/by/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work