Fast and Cost-Effective Mathematical Models for Hydrocarbon-Immiscible Water Alternating Gas Incremental Recovery...

Area of Science:

• Petroleum Engineering
• Reservoir Engineering
• Machine Learning Applications

Background:

• Enhanced Oil Recovery (EOR) is crucial for maximizing hydrocarbon extraction.
• Water-alternating-gas (WAG) injection is a proven EOR technique, typically yielding a 5-10% incremental recovery factor.
• Current reservoir modeling for WAG evaluation is time-consuming and expensive.

Purpose of the Study:

• To develop a rapid and economical mathematical model for predicting the incremental recovery factor of hydrocarbon-immiscible WAG (HC-IWAG) injection.
• To assist in designing WAG pilots and upscaling their results for medium-to-light oil in undersaturated reservoirs.
• To provide a cost-effective alternative to traditional reservoir simulation for WAG process assessment.

Main Methods:

• Literature review of WAG injection techniques.
• Development of over a thousand reservoir models for WAG and waterflood simulations.
• Application of machine learning techniques, specifically stepwise regression and Group Method of Data Handling (GMDH), for predictive model creation.

Main Results:

• Developed HC-IWAG incremental recovery factor mathematical models with a coefficient of determination (R²) of approximately 0.75.
• Identified 13 key predictor parameters influencing the WAG incremental recovery factor.
• Created interpretable and user-friendly mathematical formulas for WAG performance prediction.

Conclusions:

• The developed mathematical models offer a fast, low-cost, and accurate method for evaluating WAG EOR performance.
• These models can significantly aid subsurface teams in identifying optimal WAG candidates, optimizing injection strategies, and facilitating pilot design and upscaling.
• The research provides a practical tool for improving the efficiency and economic viability of WAG EOR projects.