Technology Focus: Flow Assurance (November 2021)

2021 ◽  
Vol 73 (11) ◽  
pp. 72-72
Author(s):  
Galen Dino
Keyword(s):  
Machine Learning ◽  
Multiphase Flow ◽  
Slug Flow ◽  
Flow Assurance ◽  
The Novel ◽  
Empirical Correlations ◽  
Liquid Holdup ◽  
Control Scheme ◽  
Technical Effort ◽  
Additional Reading

I sincerely hope that all JPT readers and your families, peers, and employers remain safe and healthy and have work as they read this year’s Flow Assurance feature. Flow-assurance effects from slug-flow engineering, design, maintenance, and operations technical concerns still create and sustain challenging technical issues requiring safe, economical solutions for both onshore unconventional and offshore conventional production facilities. The recurring long-term mitigation of slugging and various flow-assurance phenomena—along with the prevention of wax, erosion, asphaltenes, corrosion, and salt deposition—and gas hydrate prediction and handling still demand attention and considerable project technical effort. Slug-flow assessments present opportunities for significant optimization in work flows to target governing operating scenarios. Paper OTC 30172 describes an integrated iterative approach between the flow-assurance and pipeline-engineering disciplines to streamline the work flow based on the value or cost associated with changes in input parameters that affect pipeline fatigue-assessment outcomes. Case studies on two multiphase pipelines are presented to illustrate this design approach. The results show that early identification of the key pipeline profile features and dominating spans for pipeline slugging fatigue assessments facilitated the optimization of slug-flow modeling and reduced computational time. The second paper, SPE 203448, presents decision trees that are considered as nonparametric machine-learning models. The data sets used in training and testing the predictive model are experimental and were collected from literature. Air/kerosene and air/water mixtures were used in obtaining the experimental data points. Results show that the proposed boosted decision tree regression (BDTR) model outperforms the best empirical correlations and the fuzzy-logic model used in estimating liquid holdup in gas/liquid multiphase flows. For the built model, the most important input feature in estimating liquid holdup is the superficial gas velocity. The empirical correlations developed in the past for identifying liquid holdup in multiphase flow can be applied only under the flow conditions by which they were originally developed. However, this machine-learning model does not suffer from this limitation. The third paper, OTC 31298, describes a slugging-control solution that was rejected because of the use of a pseudovariable as the principal control point. A novel control scheme, therefore, was developed and tested on simulations for both hydrodynamic slugging and severe riser-induced slugging in an Angolan field. The project implemented the novel active slugging control using a topsides control valve and topsides instrumentation. While a pseudovariable, a pseudoflow controller, was used, it was part of a cascade scheme such that the principal control variable was a real top-side pressure measurement. Upon com-missioning, slugging at the facility was found to be more severe than anticipated during design, but the novel active slug-control scheme was effective in controlling incoming slugs. The desire to understand better how to describe and improve flow assurance and multiphase flow for both offshore and onshore facilities drives new production technology research, applications, and approaches. The three papers listed for additional reading focus on developing further new analytical tools while providing safe, cost-effective, and reliable operations for flow assurance. I hope you find them as interesting as I did. In addition, I invite you to join the Flow Assurance Technical Section to augment your learning. Recommended additional reading at OnePetro: www.onepetro.org. OTC 30177 - Real-Time Online Hydrate Monitoring and Prevention in Offshore Fields by Syahida Husna Azman, Petronas, et al. SPE 201316 - Modeling Dynamic Loads Induced by Slug Flows Considering Gas Expansion Caused by the Pressure Gradient in a Free Span Horizontal Hanging Pipeline by Gabriel Meneses Santos, Universidade Estadual de Campinas, et al. OTC 31238 - Taylor Bubbles of Viscous Slug Flow in Inclined Pipes by Longtong Abednego Dafyak, University of Nottingham, et al.

scholarly journals Engineering Simulation Tests on Multiphase Flow in Middle- and High-Yield Slanted Well Bores

Energies ◽  
2018 ◽  
Vol 11 (10) ◽  
pp. 2591 ◽  
Author(s):  
Dan Qi ◽  
Honglan Zou ◽  
Yunhong Ding ◽  
Wei Luo ◽  
Junzheng Yang
Keyword(s):  
Multiphase Flow ◽  
Prediction Model ◽  
Flow Pattern ◽  
Pipe Flow ◽  
Slug Flow ◽  
High Yield ◽  
Transition Flow ◽  
Liquid Holdup ◽  
Flow Pattern Map ◽  
Flow Pressure

Previous multiphase pipe flow tests have mainly been conducted in horizontal and vertical pipes, with few tests conducted on multiphase pipe flow under different inclined angles. In this study, in light of mid–high yield and highly deviated wells in the Middle East and on the basis of existent multiphase flow pressure research on well bores, multiphase pipe flow tests were conducted under different inclined angles, liquid rates, and gas rates. A pressure prediction model based on Mukherjee model, but with new coefficients and higher accuracy for well bores in the study block, was obtained. It was verified that the newly built pressure drawdown prediction model tallies better with experimental data, with an error of only 11.3%. The effect of inclination, output, and gas rate on the flow pattern, liquid holdup, and friction in the course of multiphase flow were analyzed comprehensively, and six kinds of classical flow regime maps were verified with this model. The results showed that for annular and slug flow, the Mukherjee flow pattern map had a higher accuracy of 100% and 80–100%, respectively. For transition flow, Duns and Ros flow pattern map had a higher accuracy of 46–66%.

Novel Active Slug Control in Angola - Development & Field Results

2021 ◽  
Author(s):  
Lisa Ann Brenskelle ◽  
Martin Bermudez Morles ◽  
Lauren Annette Flores
Keyword(s):  
Simulation Study ◽  
Slug Flow ◽  
Control Point ◽  
Control Variable ◽  
Mechanical Damage ◽  
Control Valve ◽  
The Novel ◽  
Control Scheme ◽  
Control Schemes ◽  
New Facility

Abstract Slug flow in multiphase flowlines can cause operational instabilities which, when severe, lead to mechanical damage or even shutdown of processing facilities. While a number of control schemes to handle slugging have been published, many of them require subsea instrumentation or make use of calculated "pseudo-variables" for control, values which have no real physical meaning. Hydrodynamic slugging was anticipated during design of a new facility in Angola, and a simulation study demonstrated that a control scheme from the published literature could be applied effectively to control the slugging. However, that solution was rejected because of the use of a pseudo-variable as the principal control point. Therefore, a novel control scheme was developed and tested on simulation for both hydrodynamic slugging and severe riser-induced slugging and later patented.(1,2) The project implemented the novel active slugging control using a topsides control valve and topsides instrumentation. While a pseudo-variable, a pseudo-flow controller, was employed, it was part of a cascade scheme such that the principal control variable was a real topside pressure measurement. Upon commissioning, slugging at the facility was found to be more severe than anticipated during design, but the novel active slug control scheme was effective in controlling incoming slugs.

scholarly journals Analysis and Prediction of COVID-19 Using SIR, SEIQR, and Machine Learning Models: Australia, Italy, and UK Cases

Information ◽  
2021 ◽  
Vol 12 (3) ◽  
pp. 109 ◽  
Author(s):  
Iman Rahimi ◽  
Amir H. Gandomi ◽  
Panagiotis G. Asteris ◽  
Fang Chen
Keyword(s):  
Machine Learning ◽  
Logistic Function ◽  
Prediction Performance ◽  
Machine Learning Algorithms ◽  
Model Parameters ◽  
The Novel ◽  
Chinese City ◽  
Limited Memory ◽  
Increasing Trend ◽  
Novel Coronavirus

The novel coronavirus disease, also known as COVID-19, is a disease outbreak that was first identified in Wuhan, a Central Chinese city. In this report, a short analysis focusing on Australia, Italy, and UK is conducted. The analysis includes confirmed and recovered cases and deaths, the growth rate in Australia compared with that in Italy and UK, and the trend of the disease in different Australian regions. Mathematical approaches based on susceptible, infected, and recovered (SIR) cases and susceptible, exposed, infected, quarantined, and recovered (SEIQR) cases models are proposed to predict epidemiology in the above-mentioned countries. Since the performance of the classic forms of SIR and SEIQR depends on parameter settings, some optimization algorithms, namely Broyden–Fletcher–Goldfarb–Shanno (BFGS), conjugate gradients (CG), limited memory bound constrained BFGS (L-BFGS-B), and Nelder–Mead, are proposed to optimize the parameters and the predictive capabilities of the SIR and SEIQR models. The results of the optimized SIR and SEIQR models were compared with those of two well-known machine learning algorithms, i.e., the Prophet algorithm and logistic function. The results demonstrate the different behaviors of these algorithms in different countries as well as the better performance of the improved SIR and SEIQR models. Moreover, the Prophet algorithm was found to provide better prediction performance than the logistic function, as well as better prediction performance for Italy and UK cases than for Australian cases. Therefore, it seems that the Prophet algorithm is suitable for data with an increasing trend in the context of a pandemic. Optimization of SIR and SEIQR model parameters yielded a significant improvement in the prediction accuracy of the models. Despite the availability of several algorithms for trend predictions in this pandemic, there is no single algorithm that would be optimal for all cases.

scholarly journals Modified Decision Tree Technique for Ransomware Detection at Runtime through API Calls

2020 ◽  
Vol 2020 ◽  
pp. 1-10
Author(s):  
Faizan Ullah ◽  
Qaisar Javaid ◽  
Abdu Salam ◽  
Masood Ahmad ◽  
Nadeem Sarwar ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Decision Tree ◽  
Feature Vector ◽  
Machine Learning Algorithms ◽  
The Novel ◽  
Proposed Model ◽  
Testing Accuracy ◽  
Financial Losses

Ransomware (RW) is a distinctive variety of malware that encrypts the files or locks the user’s system by keeping and taking their files hostage, which leads to huge financial losses to users. In this article, we propose a new model that extracts the novel features from the RW dataset and performs classification of the RW and benign files. The proposed model can detect a large number of RW from various families at runtime and scan the network, registry activities, and file system throughout the execution. API-call series was reutilized to represent the behavior-based features of RW. The technique extracts fourteen-feature vector at runtime and analyzes it by applying online machine learning algorithms to predict the RW. To validate the effectiveness and scalability, we test 78550 recent malign and benign RW and compare with the random forest and AdaBoost, and the testing accuracy is extended at 99.56%.

scholarly journals RON-Gauss: Enhancing Utility in Non-Interactive Private Data Release

2019 ◽  
Vol 2019 (1) ◽  
pp. 26-46 ◽  
Author(s):  
Thee Chanyaswad ◽  
Changchang Liu ◽  
Prateek Mittal
Keyword(s):  
Machine Learning ◽  
Real World ◽  
Differential Privacy ◽  
Real Data ◽  
The Novel ◽  
Private Data ◽  
Data Release ◽  
Machine Learning Applications ◽  
Order Of Magnitude ◽  
Real World Datasets

Abstract A key challenge facing the design of differential privacy in the non-interactive setting is to maintain the utility of the released data. To overcome this challenge, we utilize the Diaconis-Freedman-Meckes (DFM) effect, which states that most projections of high-dimensional data are nearly Gaussian. Hence, we propose the RON-Gauss model that leverages the novel combination of dimensionality reduction via random orthonormal (RON) projection and the Gaussian generative model for synthesizing differentially-private data. We analyze how RON-Gauss benefits from the DFM effect, and present multiple algorithms for a range of machine learning applications, including both unsupervised and supervised learning. Furthermore, we rigorously prove that (a) our algorithms satisfy the strong ɛ-differential privacy guarantee, and (b) RON projection can lower the level of perturbation required for differential privacy. Finally, we illustrate the effectiveness of RON-Gauss under three common machine learning applications – clustering, classification, and regression – on three large real-world datasets. Our empirical results show that (a) RON-Gauss outperforms previous approaches by up to an order of magnitude, and (b) loss in utility compared to the non-private real data is small. Thus, RON-Gauss can serve as a key enabler for real-world deployment of privacy-preserving data release.

scholarly journals Application of Various Machine Learning Techniques in Predicting Total Organic Carbon from Well Logs

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Osama Siddig ◽  
Ahmed Farid Ibrahim ◽  
Salaheldin Elkatatny
Keyword(s):  
Machine Learning ◽  
Organic Carbon ◽  
Total Organic Carbon ◽  
The Other ◽  
Well Logs ◽  
Machine Learning Techniques ◽  
Percentage Error ◽  
Average Error ◽  
Support Vector ◽  
Empirical Correlations

Unconventional resources have recently gained a lot of attention, and as a consequence, there has been an increase in research interest in predicting total organic carbon (TOC) as a crucial quality indicator. TOC is commonly measured experimentally; however, due to sampling restrictions, obtaining continuous data on TOC is difficult. Therefore, different empirical correlations for TOC have been presented. However, there are concerns about the generalization and accuracy of these correlations. In this paper, different machine learning (ML) techniques were utilized to develop models that predict TOC from well logs, including formation resistivity (FR), spontaneous potential (SP), sonic transit time (Δt), bulk density (RHOB), neutron porosity (CNP), gamma ray (GR), and spectrum logs of thorium (Th), uranium (Ur), and potassium (K). Over 1250 data points from the Devonian Duvernay shale were utilized to create and validate the model. These datasets were obtained from three wells; the first was used to train the models, while the data sets from the other two wells were utilized to test and validate them. Support vector machine (SVM), random forest (RF), and decision tree (DT) were the ML approaches tested, and their predictions were contrasted with three empirical correlations. Various AI methods’ parameters were tested to assure the best possible accuracy in terms of correlation coefficient (R) and average absolute percentage error (AAPE) between the actual and predicted TOC. The three ML methods yielded good matches; however, the RF-based model has the best performance. The RF model was able to predict the TOC for the different datasets with R values range between 0.93 and 0.99 and AAPE values less than 14%. In terms of average error, the ML-based models outperformed the other three empirical correlations. This study shows the capability and robustness of ML models to predict the total organic carbon from readily available logging data without the need for core analysis or additional well interventions.

Empirical Correlations for Prediction Slug Liquid Holdup on Slug-Pseudo-Slug and Slug-Churn Transitions in Vertical and Inclined Two-Phase Flow

2021 ◽  
pp. 1-13
Author(s):  
Ghassan H. Abdul-Majeed ◽  
Abderraouf Arabi ◽  
Gabriel Soto-Cortes
Keyword(s):  
Two Phase Flow ◽  
Liquid Velocity ◽  
Superior Performance ◽  
Phase Flow ◽  
Liquid Viscosity ◽  
Empirical Correlations ◽  
Liquid Holdup ◽  
Two Phase ◽  
Experimental Database ◽  
Transition Models

Summary Most of the existing slug (SL) to churn (CH) or SL to pseudo-slug (PS) transition models (empirical and mechanistic) account for the effect of the SL liquid holdup (HLS). For simplicity, some of these models assume a constant value of HLS in SL/CH and SL/PS flow transitions, leading to a straightforward solution. Other models correlate HLS with different flow variables, resulting in an iterative solution for predicting these transitions. Using an experimental database collected from the open literature, two empirical correlations for prediction HLS at the SL/PS and SL/CH transitions (HLST) are proposed in this study. This database is composed of 1,029 data points collected in vertical, inclined, and horizontal configurations. The first correlation is developed for medium to high liquid viscosity two-phase flow (μL > 0.01 Pa·s), whereas the second one is developed for low liquid viscosity flow (μL ≤ 0.01 Pa·s). Both correlations are shown to be a function of superficial liquid velocity (VSL), liquid viscosity (μL), and pipe inclination angle (θ). The proposed correlations in a combination with the HLS model of Abdul-Majeed and Al-Mashat (2019) have been used to predict SL/PS and SL/CH transitions, and very satisfactory results were obtained. Furthermore, the SL/CH model of Brauner and Barnea (1986) is modified by using the proposed HLST correlations, instead of using a constant value. The modification results in a significant improvement in the prediction of SL/CH and SL/PS transitions and fixes the incorrect decrease of superficial gas velocity (VSG) with increasing VSL. The modified model follows the expected increase of VSG for high VSL, shown by the published observations. The proposed combinations are compared with the existing transition models and show superior performance among all models when tested against 357 measured data from independent studies.

Artificial Intelligence and Data Analytics for Virtual Flow Metering

2021 ◽  
Author(s):  
Anton Gryzlov ◽  
Liliya Mironova ◽  
Sergey Safonov ◽  
Muhammad Arsalan
Keyword(s):  
Machine Learning ◽  
Multiphase Flow ◽  
Real Time ◽  
Production Control ◽  
Measurement Data ◽  
Sufficient Accuracy ◽  
Operating Conditions ◽  
Machine Learning Techniques ◽  
Flow Metering ◽  
Oil Gas

Abstract Modern challenges in reservoir management have recently faced new opportunities in production control and optimization strategies. These strategies in turn rely on the availability of monitoring equipment, which is used to obtain production rates in real-time with sufficient accuracy. In particular, a multiphase flow meter is a device for measuring the individual rates of oil, gas and water from a well in real-time without separating fluid phases. Currently, there are several technologies available on the market but multiphase flow meters generally incapable to handle all ranges of operating conditions with satisfactory accuracy in addition to being expensive to maintain. Virtual Flow Metering (VFM) is a mathematical technique for the indirect estimation of oil, gas and water flowrates produced from a well. This method uses more readily available data from conventional sensors, such as downhole pressure and temperature gauges, and calculates the multiphase rates by combining physical multiphase models, various measurement data and an optimization algorithm. In this work, a brief overview of the virtual metering methods is presented, which is followed by the application of several advanced machine-learning techniques for a specific case of multiphase production monitoring in a highly dynamic wellbore. The predictive capabilities of different types of machine learning instruments are explored using a model simulated production data. Also, the effect of measurement noise on the quality of estimates is considered. The presented results demonstrate that the data-driven methods are very capable to predict multiphase flow rates with sufficient accuracy and can be considered as a back-up solution for a conventional multiphase meter.

Prediction of Gas Turbine Trip: a Novel Methodology Based on Random Forest Models

2021 ◽  
Author(s):  
Enzo Losi ◽  
Mauro Venturini ◽  
Lucrezia Manservigi ◽  
Giuseppe Fabio Ceschini ◽  
Giovanni Bechini ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Gas Turbine ◽  
Gas Turbines ◽  
Remaining Useful Life ◽  
Training Data ◽  
The Novel ◽  
Novel Approach ◽  
Forest Models ◽  
Random Forest Models

Abstract A gas turbine trip is an unplanned shutdown, of which the most relevant consequences are business interruption and a reduction of equipment remaining useful life. Thus, understanding the underlying causes of gas turbine trip would allow predicting its occurrence in order to maximize gas turbine profitability and improve its availability. In the ever competitive Oil & Gas sector, data mining and machine learning are increasingly being employed to support a deeper insight and improved operation of gas turbines. Among the various machine learning tools, Random Forests are an ensemble learning method consisting of an aggregation of decision tree classifiers. This paper presents a novel methodology aimed at exploiting information embedded in the data and develops Random Forest models, aimed at predicting gas turbine trip based on information gathered during a timeframe of historical data acquired from multiple sensors. The novel approach exploits time series segmentation to increase the amount of training data, thus reducing overfitting. First, data are transformed according to a feature engineering methodology developed in a separate work by the same authors. Then, Random Forest models are trained and tested on unseen observations to demonstrate the benefits of the novel approach. The superiority of the novel approach is proved by considering two real-word case-studies, involving filed data taken during three years of operation of two fleets of Siemens gas turbines located in different regions. The novel methodology allows values of Precision, Recall and Accuracy in the range 75–85 %, thus demonstrating the industrial feasibility of the predictive methodology.

Sign in / Sign up

Export Citation Format

Share Document