Optimization of Water Alternating Gas (WAG) Injection for Enhanced Oil Recovery in Marginal Reservoirs Using Machine Learning-Driven Process Simulation

 

Table Of Contents


Chapter ONE

INTRODUCTION

  • 1.1Introduction
  • 1.2Background of Study
  • 1.3Problem Statement
  • 1.4Objective of Study
  • 1.5Limitation of Study
  • 1.6Scope of Study
  • 1.7Significance of Study
  • 1.8Structure of the Research
  • 1.9Definition of Terms

Chapter TWO

LITERATURE REVIEW

  • 2.1Review of Enhanced Oil Recovery (EOR) Methods
  • 2.2Fundamentals of Waterflooding and WAG Techniques
  • 2.3Machine Learning in Petroleum Engineering: Applications and Gaps
  • 2.4Process Simulation in Reservoir Engineering
  • 2.5Data Analytics for Reservoir Characterization
  • 2.6Numerical Modeling of Multiphase Flow
  • 2.7WAG in Marginal Reservoirs: Challenges and Opportunities
  • 2.8Reservoir Heterogeneity and Its Impact on WAG
  • 2.9Historical Case Studies of WAG Implementation
  • 2.10Gaps and Research Gaps

Chapter THREE

SYSTEM DESIGN AND IMPLEMENTATION

  • 3.1Research Philosophy and Design
  • 3.2Data Acquisition and Quality Assurance
  • 3.3Reservoir Model Selection and Characterization
  • 3.4Experimental Setup: Process Simulation Framework
  • 3.5Machine Learning Model Selection and Training
  • 3.6Feature Engineering and Data Preprocessing
  • 3.7Model Validation and Uncertainty Quantification
  • 3.8Optimization Strategy for WAG Parameters
  • 3.9Sensitivity Analysis
  • 3.10Ethical, Legal, and Safety Considerations

Chapter FOUR

SYSTEM TESTING AND EVALUATION

  • 4.1Baseline Reservoir Simulation Results
  • 4.2WAG Parameter Space Exploration
  • 4.3ML-Driven Optimization of Injection Ratios and Slug Sizes
  • 4.4Gas–Water Contact and Mobility Ratio Effects
  • 4.5Economic Evaluation and Net Present Value Analysis
  • 4.6Sensitivity to Core/Rock Properties
  • 4.7Uncertainty Propagation and Robustness of Solutions
  • 4.8Comparative Assessment: Conventional vs. ML-Driven WAG

Chapter FIVE

SUMMARY, CONCLUSION AND RECOMMENDATIONS

  • 5.1Summary of Key Findings
  • 5.2Theoretical and Practical Implications
  • 5.3Recommendations for Field Implementation
  • 5.4Limitations and Potential Biases
  • 5.5Recommendations for Future Research
  • 5.6Conclusions

Project Abstract

Optimization of Water Alternating Gas (WAG) Injection for Enhanced Oil Recovery in Marginal Reservoirs Using Machine Learning-Driven Process Simulation investigates a novel, integrated framework that combines physical reservoir engineering with data-driven modeling to maximize hydrocarbon recovery while minimizing costs and environmental impact in marginal reservoirs. The study develops a synergistic approach that leverages high-fidelity reservoir simulations, artificial intelligence, and real-time data assimilation to optimize WAG design parameters, including gas-oil ratio, slug size, injection rate, cycle duration, and polymer or surfactant additives when applicable. A comprehensive dataset is assembled from Field-ScaleOBS, historical production logs, core analyses, PVT properties, and seismic attributes to calibrate a hybrid modeling environment. This environment couples mechanistic multiphase flow simulations with machine learning surrogates to accelerate optimization loops without sacrificing physical fidelity. Key objectives include (i) quantifying the sensitivity of WAG performance to reservoir heterogeneity, mobility ratio, capillary pressure, and relative permeability curves in marginal plays; (ii) developing ML-driven process simulators that replicate critical reservoir responses with significantly reduced computation times; (iii) formulating robust optimization strategies that maximize net present value (NPV), ultimate recovery factor, and sweep efficiency under uncertainty; (iv) integrating uncertainty quantification and risk assessment to identify economically viable WAG configurations across varying production scenarios; and (v) outlining implementation pathways for field deployment, including data pipelines, model maintenance, and operational workflow alignment. Methodologically, the research advances a two-tier modeling framework a physics-based compositional simulator to capture gas-water-oil interactions, phase behavior, and transport phenomena, augmented by data-driven surrogates (e.g., deep neural networks, gradient-boosted trees) trained on high-fidelity runs to forecast key performance indicators rapidly. An adaptive optimization engine, incorporating stochastic optimization and Bayesian updating, explores the decision space of WAG cycles, injection strategies, and chemical-enhanced recovery options under parameter uncertainty. The study also investigates the role of machine learning in predicting breakthrough events, gas channeling tendencies, and reservoir-scale pressure maintenance requirements, thereby enabling proactive adjustments to injection plans. Results demonstrate improved recovery factors and economic performance relative to conventional WAG approaches, with substantial reductions in computational time enabling near real-time decision support. The research discusses scalability to field developments, data integrity challenges, and the need for rigorous validation against pilot tests and long-term production data. Sensitivity analyses reveal critical controls on performance, while the uncertainty quantification framework provides probabilistic confidence bounds for proposed operating envelopes. The work contributes a replicable methodology for leveraging machine learning-driven process simulation to optimize WAG schemes in challenging, heterogeneous marginal reservoirs, offering a pathway toward more resilient and efficient EOR practices.

Project Overview

What This Project Is About
A plain-language overview of how injecting gas and water together can help extract more oil from difficult reservoirs, using computer-driven models to predict outcomes. The project studies how to safely and efficiently mix water and gas (WAG) to push more oil to the surface, and how simple machine learning tools can improve the process simulation so decisions are faster and more reliable.

The Problem It Addresses
Many oil reservoirs are tough to produce from, with little oil left in accessible places. Traditional methods either use water or gas alone and often leave a lot of oil behind. WAG attempts to combine both to boost recovery, but choosing the best mix, timing, and conditions is complex. The project aims to find ways to predict and optimize these choices with easier-to-use computer models, saving time and resources.

Objectives of the Project


  1. Explain what WAG is and why it matters for marginal reservoirs.
  2. Learn the basics of process simulation and machine learning in this context.
  3. Develop a simple, user-friendly model to test WAG scenarios.
  4. Identify key factors that affect oil recovery and costs.
  5. Suggest practical guidelines for field deployment of optimized WAG strategies.


What You Will Do Step by Step


  1. Review basic concepts of reservoir oil recovery and WAG technique.
  2. Gather publicly available data or create representative synthetic data for simulations.
  3. Build a straightforward machine learning model to predict outcomes of WAG configurations.
  4. Integrate the ML model with a simple process simulator to run scenarios.
  5. Test different gas/water ratios, injection rates, and timings.
  6. Evaluate results by comparing oil recovery and costs.
  7. Analyze sensitivity to key inputs and identify robust operating ranges.
  8. Prepare practical recommendations and a user guide for non-experts.


Expected Outcome


A clear, easy-to-use framework that predicts the performance of WAG plans in marginal reservoirs and highlights cost-effective operating ranges. The project should deliver simple guidelines for practitioners and a demonstration that ML-driven simulations can speed up decision-making without sacrificing reliability.

Blazingprojects Mobile App

πŸ“š Over 50,000 Project Materials
πŸ“± 100% Offline: No internet needed
πŸ“ Over 98 Departments
πŸ” Software coding and Machine construction
πŸŽ“ Postgraduate/Undergraduate Research works
πŸ“₯ Instant Whatsapp/Email Delivery

Blazingprojects App

Related Research

Petroleum engineerin. 4 min read

AI-assisted optimization of enhanced oil recovery using nanofluid flooding in carbon...

What This Project Is About A straightforward exploration of using tiny, specially designed liquids (nanofluids) to push more oil out of carbonate rock formation...

BP
Blazingprojects
Read more →
Petroleum engineerin. 3 min read

Optimization of Enhanced Oil Recovery using Nano-Fluid Assisted Polymer Flooding in ...

What This Project Is About A straightforward look at how tiny fluid particles (nano-fluids) mixed with specialized polymers could help extract more oil from old...

BP
Blazingprojects
Read more →
Petroleum engineerin. 3 min read

Optimization of Water Alternating Gas (WAG) Injection for Enhanced Oil Recovery in M...

What This Project Is About A plain-language overview of how injecting gas and water together can help extract more oil from difficult reservoirs, using computer...

BP
Blazingprojects
Read more →
Petroleum engineerin. 2 min read

Nano-enhanced Oil Recovery using Hybrid Surfactant-Polymer Flooding in Ultra-Low Per...

What This Project Is About A straightforward look at using tiny particles and tailored chemicals to push more oil out of very tight carbonate rocks. The project...

BP
Blazingprojects
Read more →
Petroleum engineerin. 2 min read

Optimization of Enhanced Oil Recovery using Hybrid CO2-Nanoparticle Foam Flooding in...

What This Project Is About This project looks at ways to recover more oil from old wells by combining two approaches: using CO2 gas to push oil and tiny particl...

BP
Blazingprojects
Read more →
Petroleum engineerin. 3 min read

Smart enhanced oil recovery optimization using machine learning-guided polymer flood...

What This Project Is About A readable overview of how we can improve oil recovery by combining smart computer ideas with polymer fluids in tricky, non-uniform r...

BP
Blazingprojects
Read more →
Petroleum engineerin. 3 min read

Integrated optimization of enhanced oil recovery using nanotechnology-assisted polym...

What This Project Is About A plain-language overview of how enhanced oil recovery can be improved by combining polymer flooding with tiny, engineered particles ...

BP
Blazingprojects
Read more →
Petroleum engineerin. 4 min read

Optimization of Enhanced Oil Recovery via Polymer Flooding in Low-Permeability Reser...

What This Project Is About A simple study of how polymer flooding can help extract more oil from tight rock formations where water alone doesn’t move oil well...

BP
Blazingprojects
Read more →
WhatsApp Click here to chat with us