Artificial neural network approach for flow regime classification in gas–liquid–fiber flows based on frequency domain analysis of pressure signals

Accurate Flow Regime Classification and Void Fraction Measurement in Two-Phase Flowmeters Using Frequency-Domain Feature Extraction and Neural Networks

Two-phase flow is very important in many areas of science, engineering, and industry. Two-phase flow comprising gas and liquid phases is a common occurrence in oil and gas related industries. This study considers three flow regimes, including homogeneous, annular, and stratified regimes ranging from 5–90% of void fractions simulated via the Mont Carlo N-Particle (MCNP) Code. In the proposed model, two NaI detectors were used for recording the emitted photons of a cesium 137 source that pass through the pipe. Following that, fast Fourier transform (FFT), which aims to transfer recorded signals to frequency domain, was adopted. By analyzing signals in the frequency domain, it is possible to extract some hidden features that are not visible in the time domain analysis. Four distinctive features of registered signals, including average value, the amplitude of dominant frequency, standard deviation (STD), and skewness were extracted. These features were compared to each other to determine the best feature that can offer the best separation. Furthermore, artificial neural network (ANN) was utilized to increase the efficiency of two-phase flowmeters. Additionally, two multi-layer perceptron (MLP) neural networks were adopted for classifying the considered regimes and estimating the volumetric percentages. Applying the proposed model, the outlined flow regimes were accurately classified, resulting in volumetric percentages with a low root mean square error (RMSE) of 1.1%.

Pipeline Two-Phase Flow Pressure Drop Algorithm for Multiple Inclinations

A Generalized Additive Model (GAM) is proposed to predict the pressure drop in a gas–liquid two-phase flow at horizontal, vertical, and inclined pipes based on 21 different dimensionless numbers. It is fitted from 4605 points, considering a fluid pattern classification as Annular, Bubbly, Intermittent, and Segregated. The GAM non-parametric method reached high prediction capacity and allowed a great degree of interpretability (i.e., it helped to visualize and test statistical inference), considering that each predictor’s marginal effects could be described, unlike in other Machine Learning (ML) methods. The prediction capacity of the GAM model for the pressure gradient obtained an adjusted R2 and a mean relative error of 99.1% and 12.93%, respectively. This capacity is maintained even when ignoring Bubbly flow in the training sample. A regularization technique to filter some variables was used, but most of the predictors must maintain the model’s high predictive ability. For example, dimensionless numbers such as the Reynolds, Froude, and Weber numbers show p-values of less than 0.01% to explain the pressure gradient in the different flow patterns. The model performs adequately on 500 randomly sampled data points not used to fit the model with an error lower than 15%. The variable importance for the model and the relationship with the pressure gradient is evaluated based on the obtained splines and p-values.

Machine learning applications to predict two-phase flow patterns

Recent advances in artificial intelligence with traditional machine learning algorithms and deep learning architectures solve complex classification problems. This work presents the performance of different artificial intelligence models to classify two-phase flow patterns, showing the best alternatives for this specific classification problem using two-phase flow regimes (liquid and gas) in pipes. Flow patterns are affected by physical variables such as superficial velocity, viscosity, density, and superficial tension. They also depend on the construction characteristics of the pipe, such as the angle of inclination and the diameter. We selected 12 databases (9,029 samples) to train and test machine learning models, considering these variables that influence the flow patterns. The primary dataset is Shoham (1982), containing 5,675 samples with six different flow patterns. An extensive set of metrics validated the results obtained. The most relevant characteristics for training the models using Shoham (1982) dataset are gas and liquid superficial velocities, angle of inclination, and diameter. Regarding the algorithms, the Extra Trees model classifies the flow patterns with the highest degree of fidelity, achieving an accuracy of 98.8%.

Machine-Learning Methods for Computational Science and Engineering

The re-kindled fascination in machine learning (ML), observed over the last few decades, has also percolated into natural sciences and engineering. ML algorithms are now used in scientific computing, as well as in data-mining and processing. In this paper, we provide a review of the state-of-the-art in ML for computational science and engineering. We discuss ways of using ML to speed up or improve the quality of simulation techniques such as computational fluid dynamics, molecular dynamics, and structural analysis. We explore the ability of ML to produce computationally efficient surrogate models of physical applications that circumvent the need for the more expensive simulation techniques entirely. We also discuss how ML can be used to process large amounts of data, using as examples many different scientific fields, such as engineering, medicine, astronomy and computing. Finally, we review how ML has been used to create more realistic and responsive virtual reality applications.

Data driven methodology for model selection in flow pattern prediction

The determination of multiphase flow parameters such as flow pattern, pressure drop and liquid holdup, is a very challenging and valuable problem in chemical, oil and gas industries, especially during transportation. There are two main approaches to solve this problem in literature: data based algorithms and mechanistic models. Although data based methods may achieve better prediction accuracy, they fail to explain the two-phase characteristics (i.e. pressure gradient, holdup, gas and liquid local velocities, etc.). Recently, many approaches have been made for establishing a unified mechanistic model for steady-state two-phase flow to predict accurately the mentioned properties. This paper proposes a novel data-driven methodology for selecting closure relationships from the models included in the unified model. A decision tree based model is built based on a data driven methodology developed from a 27670 points data set and later tested for flow pattern prediction in a set made of 9224 observations. The closure relationship selection model achieved high accuracy in classifying flow regimes for a wide range of two-phase flow conditions. Intermittent flow registering the highest accuracy (86.32%) and annular flow the lowest (49.11%). The results show that less than 10% of global accuracy is lost compared to direct data based algorithms, which is explained by the worse performance presented for atypical values and zones close to boundaries between flow patterns.