Artificial Intelligence Techniques for the Hydrodynamic Characterization of Two-Phase Liquid–Gas Flows: An Overview and Bibliometric Analysis DOI Creative Commons
July Andrea Gómez Camperos,

Marlon Mauricio Hernández Cely,

Aldo Pardo García

et al.

Fluids, Journal Year: 2024, Volume and Issue: 9(7), P. 158 - 158

Published: July 8, 2024

Accurately and instantly estimating the hydrodynamic characteristics in two-phase liquid–gas flow is crucial for industries like oil, gas, other multiphase sectors to reduce costs emissions, boost efficiency, enhance operational safety. This type of involves constant slippage between gas liquid phases caused by a deformable interface, resulting changes volumetric fraction creation structures known as patterns. Empirical numerical methods used prediction often result significant inaccuracies during scale-up processes. Different methodologies based on artificial intelligence (AI) are currently being applied predict flow, which was corroborated with bibliometric analysis where AI techniques were found have been pattern recognition, determination each fluid, pressure gradient estimation. The results revealed that total 178 keywords 70 articles, 29 reached threshold (machine learning, pattern, intelligence, neural networks high predominance), published mainly Flow Measurement Instrumentation. journal has highest number articles related studied topic, nine articles. most relevant author Efteknari-Zadeh, E, from Institute Optics Quantum Electronics.

Language: Английский

Application of artificial neural network to multiphase flow metering: A review DOI

Siamak Bahrami,

Saeid Alamdari,

Mohammadreza Farajmashaei

et al.

Flow Measurement and Instrumentation, Journal Year: 2024, Volume and Issue: 97, P. 102601 - 102601

Published: April 26, 2024

Language: Английский

Citations

13
An optimised and novel capacitance-based sensor design for measuring void fraction in gas–oil two-phase flow systems DOI Creative Commons
Abdullah M. Iliyasu, Mohammad Hossein Shahsavari,

Abdullah S. Benselama

et al.

Nondestructive Testing And Evaluation, Journal Year: 2024, Volume and Issue: 39(8), P. 2450 - 2466

Published: Jan. 9, 2024

This study explores a new electrode configuration for measuring the void fraction of two-phase flows using capacitance-based sensors. The proposed method is considered 'skewed' because its unique geometric shape, and performance sensor was evaluated improved via multiple simulations COMSOL Multiphysics software. encompass three different flow patterns, stratified, annular homogeneous, whose themselves were verified in previous study. influences properties parameters on sensitivity to determine an optimal configuration. Furthermore, distribution fractions analysed various patterns. Additionally, also compared alongside double-ring concave sensors overall sensitivity. At 2.11 pF, significantly higher than that other It worth mentioning measurement precision multiphase meters high importance, particularly petroleum industry oil price amount transported products.

Language: Английский

Citations

6
Measuring volume fractions of a three-phase flow without separation utilizing an approach based on artificial intelligence and capacitive sensors DOI Creative Commons
Abdulilah Mohammad Mayet, Farhad Fouladinia, Seyed Mehdi Alizadeh

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0301437 - e0301437

Published: May 16, 2024

Many different kind of fluids in a wide variety industries exist, such as two-phase and three-phase. Various combinations them can be expected gas-oil-water is one the most common flows. Measuring volume fraction phases without separation vital many aspects, which financial issues. methods are utilized to ascertain volumetric proportion each phase. Sensors based on measuring capacity so popular because this sensor operates seamlessly autonomously necessitating any form segregation or disruption for process. Besides, at present moment, Artificial intelligence (AI) nominated useful tool several fields, metering no exception. Also, three main type regimes found annular, stratified, homogeneous. In paper, fractions three-phase homogeneous regime measured. To accomplish objective, an Neural Network (ANN) capacitance-based utilized. train presented network, optimized was implemented COMSOL Multiphysics software after doing lot simulations, 231 data produced. Among all obtained results, 70 percent (161 data) awarded data, rest (70 considered test data. This investigation proposes new intelligent system Multilayer Perceptron network (MLP) that estimate water-oil-gas fluid’s water precisely with very low error. The Mean Absolute Error (MAE) equal 1.66. dedicates predicting method’s considerable accuracy. Moreover, study confined cannot measure void other fluid types future works. temperature pressure changes highly temper relative permittivity density liquid inside pipe another idea.

Language: Английский

Citations

6
A Novel Smart Optimized Capacitance-Based Sensor for Annular Two-Phase Flow Metering With High Sensitivity DOI Creative Commons
Rahmad Syah, Aryan Veisi, Zainal A. Hasibuan

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 60709 - 60716

Published: Jan. 1, 2023

Accurately determining phase fractions in two-phase flows is among the most significant issues Industries related to production and processing of petroleum petrochemicals. There are numerous sensor types configurations for measuring void fraction. In this respect, capacitance-based commonly recognized as one precise widely utilized sensors. essay, COMSOL Multiphysics software, which has been benchmarked, was used simulations with various electrode architectures oil-air flow an annular pattern. The initial were helix, double ring, concave parallel plates. Finite element analysis utilizing executed compare configurations. Results exposed disparate sensitivities different geometries. To get better results, a new geometry called arrow-shaped optimized Artificial intelligence (AI) proposed compared others. responses presented demonstrated that had 21% higher sensitivity than best-performing four other existing designs, including concave, These results indicate superior performance its potential use high-sensitivity applications.

Language: Английский

Citations

14
An artificial neural network and a combined capacitive sensor for measuring the void fraction independent of temperature and pressure changes for a two-phase homogeneous fluid DOI
Abdulilah Mohammad Mayet,

Gorelkina Evgeniya Ilyinichna,

Farhad Fouladinia

et al.

Flow Measurement and Instrumentation, Journal Year: 2023, Volume and Issue: 93, P. 102406 - 102406

Published: June 22, 2023

Language: Английский

Citations

11
A novel metering system consists of capacitance-based sensor, gamma-ray sensor and ANN for measuring volume fractions of three-phase homogeneous flows DOI
Farhad Fouladinia, Seyed Mehdi Alizadeh,

Evgeniya Ilyinichna Gorelkina

et al.

Nondestructive Testing And Evaluation, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 27

Published: July 7, 2024

Measuring the volume fraction of different types fluids with two or three phases is so vital. Among all available methods, them, capacitance-based and gamma-ray attenuation, are popular widely used. Moreover, nowadays, AI which stands for Artificial Intelligence can be seen almost in areas, measuring section no exception. In this paper, main goal to predict a three-phase homogeneous fluid contains water, oil, gas materials. To opt an optimised method, combination sensors, attenuation sensor Neural Networks (ANN) utilised. train proposed metering system MLP type, inputs considered. For first input, concave simulated COMSOL Multiphysics software combinations (different fractions) applied. Then through theoretical investigations sensor, Barium-133 radiates 0.356 MeV This way, second required input generated. Finally, implement new accurate system, number networks characteristics run MATLAB software. The best structure had Mean Absolute Error (MAE) equal 0.33, 3.68 3.75 oil phases, respectively. accuracy presented illustrated by received outcomes. novelty study proposing combined method that measure fluid's fractions containing precisely.

Language: Английский

Citations

4
Metrologically Interpretable Soft-Sensing Technique for Non-Invasive Liquid Flow Estimation from Vibration Data DOI Creative Commons
Gabriel Thaler, João Paulo Zomer Machado, Rodolfo C.C. Flesch

et al.

Metrology, Journal Year: 2025, Volume and Issue: 5(1), P. 6 - 6

Published: Jan. 15, 2025

This paper proposes a metrologically interpretable soft sensing method for estimating the liquid flow rates in hydraulic systems from non-invasive vibration frequency power band data. Despite considerable interest estimation, state-of-the-art methods provide little to no metrological capabilities. In this work, dedicated test rig was developed automatically acquire and rate data centrifugal pump, range between 0.05 × 10−5m3/s 9.11 10−5m3/s. The were processed into bands, which subsequently used optimize train multilayer perceptron neural network sensing. trained model compared with models different processing literature. resulted root mean squared error 75.4% smaller than second-best cross-validation, 51.5% uncertainty of proposed regression estimated using combination ensemble learning Monte Carlo simulations, combined reference sensor obtain total sensor, found be 3.9 10−6m3/s 6.1 throughout measured range. accuracy largest individual contribution final uncertainty, closely followed by uncertainty.

Language: Английский

Citations

0
Application of selected methods of computational intelligence to recognition of the liquid–gas flow regime in pipeline by use gamma absorption and frequency domain feature extraction DOI
Robert Hanus, Marcin Zych, Maciej Kusy

et al.

Measurement, Journal Year: 2024, Volume and Issue: 238, P. 115260 - 115260

Published: July 18, 2024

Language: Английский

Citations

3
Intelligent Measuring of the Volume Fraction Considering Temperature Changes and Independent Pressure Variations for a Two-Phase Homogeneous Fluid Using an 8-Electrode Sensor and an ANN DOI Creative Commons
Ramy Mohammed Aiesh Qaisi, Farhad Fouladinia, Abdulilah Mohammad Mayet

et al.

Sensors, Journal Year: 2023, Volume and Issue: 23(15), P. 6959 - 6959

Published: Aug. 5, 2023

Two-phase fluids are widely utilized in some industries, such as petrochemical, oil, water, and so on. Each phase, liquid gas, needs to be measured. The measuring of the void fraction is vital many industries because there two-phase with a wide variety liquids. A number methods exist for fraction, most popular capacitance-based sensors. Aside from being easy use, sensor does not need any separation or interruption measure fraction. In addition, contemporary era, thanks Artificial Neural Networks (ANN), measurement have become much more accurate. same can said this paper, new metering system utilizing an 8-electrode Multilayer Perceptron network (MLP) presented predict air water volume fractions homogeneous fluid. Some characteristics, temperature, pressure, etc., impact on results obtained aforementioned sensor. Thus, considering temperature changes, proposed predicts independent pressure variations. All simulations were performed using COMSOL Multiphysics software changes 275 370 degrees Kelvin. range 1 500 Bars, was considered pressure. has inputs mentioned software, along temperature. only output belongs predicted which low MAE equal 0.38. based result, it that precisely measures amount

Language: Английский

Citations

9
Identification of the Structure of Liquid–Gas Flow in a Horizontal Pipeline Using the Gamma-Ray Absorption and a Convolutional Neural Network DOI Creative Commons
Robert Hanus, Marcin Zych,

Piotr Ochał

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(11), P. 4854 - 4854

Published: June 4, 2024

Knowledge of the liquid–gas flow regime is important for proper control many industrial processes (e.g., in mining, nuclear, petrochemical, and environmental industries). The latest publications this field concern use computational intelligence methods structure recognition, which include, example, expert systems artificial neural networks. Generally, machine learning exploit various characteristics sensors signals value, time, frequency, time–frequency domain. In work, convolutional network (CNN) VGG-16 applied analysis histogram images obtained water–air by using gamma-ray absorption. experiments were carried out on laboratory hydraulic installation fitted with a radiometric measurement system. essential part horizontal pipeline made metalplex, 4.5 m long, an internal diameter 30 mm. set used investigation consists linear Am-241 radiation source energy 59.5 keV scintillation detector NaI(Tl) crystal. four types regimes (plug, slug, bubble, transitional plug–bubble) studied. MATLAB 2022a software was to analyze signal from detector. It found that CNN correctly recognizes more than 90% cases.

Language: Английский

Citations

2