Deep-Learning-Based Approach for Optimizing Infill Well Placement

Abstract This study introduces a novel machine learning-based framework aimed at optimizing well placement for shale development. Traditional well-placement techniques, often reliant on physics-based modeling or empirical assumptions, have proven to be time-consuming and prone to suboptimal results. To address these limitations, we propose a Deep Convolutional Neural Networks (DCNN) based workflow that leverages context-specific algorithms for accurate subsurface-driven infill planning optimization. The framework was validated on real-world data from 2 fields of 630 wells in the Permian basin, successfully replicating the validation dataset with 94% accuracy and reducing the time required for analysis by over 85% compared to conventional methods. Furthermore, the framework demonstrated improved Estimated Ultimate Recovery (EUR) and effective reserve management, emphasizing its potential to enhance the economic viability of shale production.