product iamge

Machine Learning and Flow Assurance in Oil and Gas Production

This document was uploaded by one of our users. The uploader already confirmed that they had the permission to publish it. If you are author/publisher or own the copyright of this documents, please report to us by using this DMCA report form.

Download
Search on Amazon

Simply click on the Download Book button.

Yes, Book downloads on Ebookily are 100% Free.

Sometimes the book is free on Amazon As well, so go ahead and hit "Search on Amazon"

This book is useful to flow assurance engineers, students, and industries who wish to be flow assurance authorities in the twenty-first-century oil and gas industry.

The use of digital or artificial intelligence methods in flow assurance has increased recently to achieve fast results without any thorough training effectively. Generally, flow assurance covers all risks associated with maintaining the flow of oil and gas during any stage in the petroleum industry. Flow assurance in the oil and gas industry covers the anticipation, limitation, and/or prevention of hydrates, wax, asphaltenes, scale, and corrosion during operation. Flow assurance challenges mostly lead to stoppage of production or plugs, damage to pipelines or production facilities, economic losses, and in severe cases blowouts and loss of human lives. A combination of several chemical and non-chemical techniques is mostly used to prevent flow assurance issues in the industry. However, the use of models to anticipate, limit, and/or prevent flow assurance problems is recommended as the best and most suitable practice. The existing proposed flow assurance models on hydrates, wax, asphaltenes, scale, and corrosion management are challenged with accuracy and precision. They are not also limited by several parametric assumptions. Recently, machine learning methods have gained much attention as best practices for predicting flow assurance issues. Examples of these machine learning models include conventional approaches such as artificial neural network, support vector machine (SVM), least square support vector machine (LSSVM), random forest (RF), and hybrid models. The use of machine learning in flow assurance is growing, and thus, relevant knowledge and guidelines on their application methods and effectiveness are needed for academic, industrial, and research purposes.

In this book, the authors focus on the use and abilities of various machine learning methods in flow assurance. Initially, basic definitions and use of machine learning in flow assurance are discussed in a broader scope within the oil and gas industry. The rest of the chapters discuss the use of machine learning in various flow assurance areas such as hydrates, wax, asphaltenes, scale, and corrosion. Also, the use of machine learning in practical field applications is discussed to understand the practical use of machine learning in flow assurance.

Author(s): Bhajan Lal, Cornelius Borecho Bavoh, Jai Krishna Sahith Sayani
Publisher: Springer
Year: 2023

Language: English
Pages: 178
City: Cham

Preface
Contents
1 Machine Learning and Flow Assurance Issues
1.1 Introduction
1.2 Flow Assurances Challenges
1.2.1 Wax Deposition
1.2.2 Corrosion
1.2.3 Asphaltene
1.2.4 Scales
1.2.5 Hydrates
1.3 Machine Learning Vocabulary
References
2 Machine Learning in Oil and Gas Industry
2.1 Introduction
2.2 Machine Learning in Upstream
2.2.1 Exploration
2.2.2 Reservoir Engineering
2.2.3 Drilling Engineering
2.2.4 Production Engineering
2.3 Machine Learning Advancements in the Oil and Gas Industry
2.3.1 Total S.A. With Google Cloud–Optimize Subsurface Data Analysis
2.3.2 ExxonMobil and MIT Collaborate to Detect Oil Seeps with AI-Powered Robots
2.3.3 Shell—Machine Learning Algorithms for Precision Drilling
2.3.4 Aker BP and Spark Cognition—Predictive Maintenance Increases Productivity
2.4 Challenges
2.4.1 Manpower
2.4.2 Data Availability
2.4.3 Opportunities and Facilities for Collaboration
2.5 COVID-19's Impact on the Oil and Gas Industry, and AI as a Solution Companies
2.6 Summary
References
3 Multiphase Flow Systems and Potential of Machine Learning Approaches in Cutting Transport and Liquid Loading Scenarios
3.1 Introduction to Multiphase
3.1.1 Gas–Liquid Flow Systems
3.1.2 Liquid–Liquid Flow Systems
3.1.3 Solid–Liquid Flow Systems
3.1.4 Solid–Liquid–Gas Three-Phase Flow Systems
3.2 Flow Assurance Issues in Drilling Applications (Cutting Transport)
3.3 Introduction of Cutting Transport Issues
3.4 Evolution of Various Cutting Transport Models
3.4.1 Layer Model
3.4.2 Layer Model
3.5 Empirical Model
3.6 Transient Model
3.7 Machine Learning Approaches for Cutting Transport
3.8 Flow Assurance issues in Liquid Loading Applications
3.8.1 Introduction of Liquid Loading Issue
3.8.2 Flow Pattern Analysis for Liquid Loading System
3.8.3 Prediction Models for Liquid Loading
3.8.4 Machine Learning Approaches for Liquid Loading or Gas/liquid Flow
3.9 Case Studies in Multiphase Flow Assurance
3.9.1 Conclusion
References
4 Machine Learning in Corrosion
4.1 Introduction
4.2 Corrosion in Oil and Gas Industry
4.2.1 Corrosion Mechanism
4.2.2 Corrosion Factors
4.2.3 Types of Corrosion
4.2.4 Corrosion Control
4.3 Mitigation Procedures
4.3.1 Pigging
4.3.2 Corrosion Inhibitor
4.3.3 Internal Coating
4.3.4 External Coating
4.3.5 Cathodic Protection
4.3.6 Process Optimization
4.4 Corrosion Prediction Models
4.4.1 Hydrocor
4.4.2 Cassandra
4.4.3 De Waard
4.4.4 NORSOK
4.4.5 Lipucor
4.4.6 ECE
4.4.7 KSC
4.5 Machine Learning in Corrosion
4.5.1 Weight Loss Method
4.5.2 Tafel Extrapolation Method
4.5.3 Machine Learning
4.6 Case Study
References
5 Machine Learning in Asphaltenes Mitigation
5.1 Introduction
5.2 Asphaltene Precipitation and Deposition in Oil and Gas Industry
5.3 Asphaltene Mitigation Procedures
5.3.1 Mechanical Method
5.3.2 Ultrasonic Treatment
5.3.3 Thermal Treatment
5.3.4 Bacterial Treatment
5.3.5 Chemical Treatment
5.4 Asphaltene Prediction Models
5.4.1 Thermodynamic Solubility Technique
5.4.2 Colloidal Technique
5.4.3 Asphaltene Deposition Modelling
5.5 Machine Learning Application in Asphaltenes Precipitation and Deposition Control
5.5.1 Case Study
5.6 Conclusion
References
6 Machine Learning for Scale Deposition in Oil and Gas Industry
6.1 Introduction
6.2 Source of Scaling in Oil and Gas Industry
6.3 Mechanism of Scale Deposition
6.3.1 Types of Scales
6.3.2 Influence of Impurities on Scale Formation
6.3.3 Scale Control Methods
6.4 Effect of Scaling to Equipment Pipelines
6.5 Scale Inhibition Placement
6.5.1 Scale Inhibition Placement by the Squeeze Technique
6.5.2 Pumping the Inhibitor with the Stimulation
6.5.3 Pumping the Inhibitor with Fracturing Fluid
6.5.4 Inhibitor Impregnated into Proppant
6.6 Prediction Models Available for Scale Formation Detection
6.6.1 Saturation Index
6.6.2 Inhibitor Impregnated into Proppant
6.6.3 Ion Association Theory
6.7 Machine Learning for Scale Deposition
6.7.1 K-Nearest Neighbor
6.7.2 Gradient Boosting Classifier
6.7.3 Decision Tree
6.7.4 Support Vector Machines (SVM)
6.7.5 Gradient Boosting
6.8 Applications of Artificial Intelligence in Oil Desalination Systems
6.8.1 FN Tool to Develop Scale-Formation Correlation
6.8.2 Least Square Support Vector Machine (LSSVM)
6.8.3 Scale Thickness Measurement Using Gamma-Ray Densitometer
6.8.4 Prediction of CaCO3 Scaling Using MLP and PNN
6.8.5 Prediction of BaSO4/CaSO4 Oilfield Scale Using ANN and SNN
6.9 Case Studies on Scaling Measurement Using Machine Learning
6.9.1 Effective Control of Deposition of Scale Using AI in Vacuum Pump
6.9.2 Conclusion
References
7 Machine Learning in CO2 Sequestration
7.1 Introduction
7.2 Conventional CO2 Sequestration Techniques
7.2.1 Enhanced Oil Recovery
7.2.2 Depleted Reservoirs
7.2.3 Deep Saline Aquifers
7.2.4 Unmineable Coal Seams
7.2.5 Conventional/Prediction Models for CO2 Sequestration
7.3 Machine Learning in CO2 Sequestration
7.3.1 Machine Learning in Enhanced Oil Recovery (EOR)
7.3.2 Machine Learning in Saline Aquifers
7.3.3 Machine Learning in Depleted Reservoirs
7.4 Conclusion
References
8 Machine Learning in Wax Deposition
8.1 Introduction
8.1.1 Wax Deposition in Oil and Gas Industry
8.2 Wax Deposition Mitigation Techniques
8.2.1 Mechanical Techniques
8.2.2 Heating
8.2.3 Chemical Inhibitors
8.2.4 Microbiological techniques
8.3 Prediction’s Models in Wax Deposition
8.4 Machine Learning in Wax Deposition
8.5 Wax Deposition Case Studies
8.6 Conclusion
References
9 Machine Learning Application in Gas Hydrates
9.1 Introduction to Gas Hydrates
9.2 Introduction to Gas Hydrates
9.2.1 Methane Recovery
9.2.2 Energy Storage
9.2.3 Desalination
9.2.4 Greenhouse Gas Capture
9.3 Conventional Gas Hydrate Mitigation Method
9.4 Chemical Inhibition of Gas Hydrates
9.4.1 Thermodynamic Hydrate Inhibitors (THIs)
9.4.2 Low Dosage Hydrate Inhibitors (LDHIs)
9.5 Flow Assurance Challenge
9.6 Machine Learning in Gas Hydrates
9.7 Case Study
9.8 Conclusion
References
10 Machine Learning Application Guidelines in Flow Assurance
10.1 Introduction
10.2 Data Selection
10.3 Data Representation
10.4 Model Development
References