A hybrid deep learning network for tight and shale reservoir characterization using pressure and rate transient data
Bibliographic record
Abstract
ABSTRACT Pressure and rate transient analyses have been widely used to determine the properties of a reservoir, such as boundary types, permeability, and formation pressure. Typically, one analytical model must be selected first to interpret the transient data; however, it is very difficult to determine which model suits the reservoir conditions. In this study, a hybrid deep learning (DL) model is developed to offer an alternative approach to characterize the reservoir (i.e., reservoir shape, flow boundary type, and porosity system) using pressure and rate transient data interpretation in unconventional tight and shale reservoirs. A multivariate synthetic data set of pressure and rate transients is first generated from the analytical solutions. Five DL architectures of recurrent, convolutional, and hybrid recurrent-convolutional types are trained on a multivariate time series of five features: pressure, pressure derivative, normalized rate, normalized cumulative production, and normalized integral derivative cumulative production. The performances of these architectures are compared, and the hybrid convolutional neural networks–long short-term memory leads to the highest performance with an accuracy of 0.98 on validation data. The performance of the optimum architecture is examined on transient data generated from reservoir simulations, and results demonstrate that the developed architecture can correctly characterize the reservoir with a accuracy of 98%. The applicability of the hybrid architecture on different length and resolution time series is also tested. It is found that the short-time flow regimes of wellbore storage and initial transition do not affect the machine learning architecture’s classification performance, and the boundary-dominated flow is the main regime for classification of these types of resources.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".