Deep Learning-Based Production Forecasting for Liquid-Rich Gas in the Duvernay Shale Play
Bibliographic record
Abstract
Abstract The Duvernay Formation is one of the most significant unconventional hydrocarbon formations in the Western Canada Sedimentary Basin (WCSB), known for its high liquid hydrocarbon content. Due to hydraulic fracturing being widely applied, the significant reservoir heterogeneity makes forecasting the newly developed well extremely challenging compared to traditional methods. Our previous work successfully applied a deep learning-based production forecasting model to the Montney shale gas play. However, Duvernay shale play exhibits significant variability in gas and liquid production proportions across different regions. This variation introduces challenges in accurately predicting multi-phase flow production behaviour. This study enhances our previously developed Masked Encoding and Decoding (MED) architecture for forecasting multi-phase hydrocarbon production from the Duvernay Formation. To mitigate the accumulation of errors typically encountered in recursive generation methods for the three production phases (oil, gas, and water), the model adopts a Non-Autoregressive Generation (NAG) approach, which predicts future production in a single step. The model integrates geostatic properties and continuously updates as new production data becomes available. Experiments were conducted using a dataset of 2,700 wells from the Duvernay Formation, with oil, gas, and water production rates pre-processed using a novel Arp's decline denoising method to enhance model stability during training. Results demonstrate the enhanced MED model's superior accuracy compared to other well-known sequence-to-sequence models, effectively capturing complex gas-liquid ratio variability and dynamically updating predictions with new data.
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 | 0.001 | 0.001 |
| 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".