Physics Inspired Machine Learning for Reliable Production Forecast in Unconventional Reservoirs
Bibliographic record
Abstract
Abstract Reliable production forecasting in unconventional reservoirs requires consideration of the underlying physics that govern subsurface flow dynamics. For unconventional shale reservoirs, the identification of flow regimes, including linear and boundary-dominated flow, provides important insights for production forecasts. Traditional rate transient analysis methods, however, often rely on manual processes, introducing a degree of subjectivity and potential bias into the results. We introduce an innovative machine learning-driven approach, rooted in the fundamental physics of flow within hydraulically fractured tight reservoirs. This approach enhances efficiency, flexibility, and automation through machine learning. It also boosts the reliability and insights in production forecasts by leveraging a robust physics-based foundation. Our workflow is constructed upon analytical solutions for multi-stage fractured shale reservoirs, assuming uniform bi-wing planar fractures and reservoir homogeneity. This simplification represents an asymptotic solution to unconventional wells and aligns with characteristic plots of field production. The first component of the workflow is to automatically analyze production data and generate characteristic attributes for linear flow and boundary-dominated flow. Following this, we employ a Markov chain Monte Carlo process that integrates actual production data with flow regime analysis, resulting in probabilistic multi-segment decline models for production forecasting with uncertainty ranges and confidence estimation. Building on these characteristics and production forecasts derived from existing producing wells, we develop a two-step machine learning model to predict future planned wells. Field applications in both the Permian Basin and Eagle Ford have demonstrated the efficiency and reliability of our proposed workflow. Operating in a fully autonomous mode, our methodology delivers results that closely align with detailed engineering forecasts for assets at various stages of development. In a fully autonomous mode, our results closely match detailed engineering forecasts when using limited data for validation testing. Additionally, the workflow is designed to be adaptable and flexible, corresponding to data quality and availability as well as the practical business needs. This innovative workflow underscores the powerful synergy between machine learning and fundamental physics in delivering efficient, reliable, and insightful solutions for engineering tasks.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".