Visualization experiment of gas–liquid flow pattern downstream of single-orifice plates in horizontal pipes under an intermittent upstream flow

Ma, Yan; Liu, Weiye; Wu, Hengliang; Liu, Yuan; Lyu, Junfu; Cai, Zhenqi

doi:10.1016/j.expthermflusci.2020.110206

Cited by 6 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The terms described above are processed by a transfer function, which precisely determines the output that is being searched for. For this study, the hyperbolic sigmoid tangent (TanSig) and logarithmic sigmoid (LogSig) activation functions were used [15], these being the most common activation functions in data processing with non-linear equations defined by equations (2) and 3 (2)…”

Section: Design Of the Artificial Neuronal Networkmentioning

confidence: 99%

“…The oil industry has focused its interest on the development of technologies that allow obtaining updated systems for the precise measurement of multiphase flow, defining this as a concurrent flow of substances in certain states or phases (solid, liquid, gas) which generate a layer separation with mixture between the phases [1] or characteristic patterns, derived from the initial hydrodynamic parameters of the Flow [2]. Hydrodynamic parameters are identified by applying different methodologies such as electrical impedance [3], pressure variation [4], ultrasonic echoes [5] and optical image analysis [6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predictive Modeling of Holdup in Horizontal Wateroil Flow Using a Neural Network Approach

Díaz¹,

González‐Estrada²,

Cely³

2021

14th WCCM-ECCOMAS Congress

View full text Add to dashboard Cite

In this work, the application of an artificial neural network (ANN) is proposed to develop a predicting model for the holdup of a two-phase flow composed of water and mineral oil in a horizontal pipe. For this, the surface velocities of each fluid and the differential pressure in the pipeline are used as input parameters of the multilayer artificial neural network with backpropagation, while the holdup of the fluids is used as the output parameter for the training. A set of 56 experimental data was obtained in the LabPetro-CEPETRO-UNICAMP laboratory. The best performing results for the predictive model show a mean absolute error (AAPE) of 3.01% and a coefficient of determination 2 of 0.9964 using 15 neurons in the hidden layer of the network and the TanSig transfer function.

show abstract

Section: Design Of the Artificial Neuronal Networkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Predictive Modeling of Holdup in Horizontal Wateroil Flow Using a Neural Network Approach

Díaz¹,

González‐Estrada²,

Cely³

2021

14th WCCM-ECCOMAS Congress

View full text Add to dashboard Cite

show abstract

“…The extraction of these substances generates the formation of flows with more than one phase consecutively which are monitored with intrusive devices or techniques [9][10][11][12] that quantify parameters of each of the fluids present [13] such as viscosities, densities, surface velocities [14,15], Reynolds number, holdups [16] and in certain cases, they are associated with geometric variables of the pipes, such as their diameter, length, and inclination [17,18]. The identification of flow patterns in pipelines is a fundamental factor [19], being of vital importance to obtain with precision a tool capable of modelling such behaviors of multiphase flows working together with existing artificial intelligence techniques [20,21] such as artificial neural networks [22], in order to train, validate and test an intelligent model [23] capable of generating flow maps [24] for vertical [25], horizontal [26] pipes in which all the performances generated from the modification of the physical…”

Section: Introductionmentioning

confidence: 99%

Flow pattern identification of liquid-liquid (oil and water) in vertical pipelines using machine learning techniques

Ruiz-Diaz

Camperos

Hernández-Cely

2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Given the importance of process control in the petrochemical industry, there is a need to determine the behavior of the fluids inside the pipes. In this work a methodology is developed for the identification of flow patterns in vertical pipes with diameters between 0.01 m and 0.10 m, from the implementation of artificial intelligence techniques, for a liquid combination of two phases composed of oil with viscosity in the range of 792 Kg/m3 to 1823 Kg/m3 and water at room temperature. The predictive models generated in the structuring of the methodology were trained with 70% of data based on viscosity parameters, pipe diameter, volume fraction and surface velocities of the working fluids stored in a database. The remaining information, equivalent to 30% of the total, was used to develop the automatic model validation. The flow patterns identified by the intelligent system for oil and water flow, without considering the predominant substance, are churning, dispersed, very fine dispersion, transition flow, intermittent, and annular

show abstract

“…La industria petrolera ha centrado su interés en el desarrollo de tecnologías enfocadas en sistemas actualizados para la medición precisa de flujo multifásico [1]; este se define como un flujo continuo de sustancias en determinados estados termodinámicos (sólido, líquido, gas), los cuales generan una capa de separación con mezcla entre las fases [2] o patrones característicos [3], [4], [5], [6] derivados de los parámetros fluidodinámicos iniciales del flujo en la vertical [7], [8] y en la horizontal [9], [10]. Los parámetros hidrodinámicos son identificados mediante la aplicación de distintas metodologías, tales como la impedancia eléctrica [11], variación de presión [12], ecos ultrasónicos [13] y análisis de imágenes ópticas [14].…”

Section: Introductionunclassified

Modelo predictivo para el cálculo de la fracción volumétrica de un flujo bifásico agua-aceite en la horizontal utilizando una red neuronal artificial

Hernández-Cely¹,

Ruiz-Diaz²,

González‐Estrada

2022

Rev. UIS ing.

View full text Add to dashboard Cite

Este artículo presenta la aplicación de una red neuronal artificial (RNA) para el desarrollo de un modelo capaz de predecir la fracción volumétrica de un flujo bifásico compuesto por agua y aceite mineral en una tubería horizontal. Se utilizan las velocidades superficiales de cada fluido y el diferencial de presión en la tubería como parámetros de entrada de la red neuronal artificial multicapa con retropropagación, mientras que la fracción volumétrica de los fluidos se utiliza como parámetro de salida en el entrenamiento de la misma. Los 56 datos experimentales con los que se trabajó se obtuvieron en el laboratorio LabPetro - CEPETRO-UNICAMP. Los resultados que arrojó el modelo predictivo con mejor rendimiento presentan un error absoluto medio porcentual (AAPE) de 3,01 % y un coeficiente de determinación de 0,9964 utilizando 15 neuronas en la capa oculta de la red y la función de transferencia TanSig.

show abstract

Visualization experiment of gas–liquid flow pattern downstream of single-orifice plates in horizontal pipes under an intermittent upstream flow

Cited by 6 publications

References 22 publications

Predictive Modeling of Holdup in Horizontal Wateroil Flow Using a Neural Network Approach

Predictive Modeling of Holdup in Horizontal Wateroil Flow Using a Neural Network Approach

Flow pattern identification of liquid-liquid (oil and water) in vertical pipelines using machine learning techniques

Modelo predictivo para el cálculo de la fracción volumétrica de un flujo bifásico agua-aceite en la horizontal utilizando una red neuronal artificial

Contact Info

Product

Resources

About