Bibliographic record
Abstract
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 166380, ’Optimized Well Modeling of Liquids-Rich Shale Reservoirs,’ by Aleksander O. Juell, SPE, and Curtis H. Whitson, SPE, NTNU/Petroleum Engineering Reservoir Analysts, prepared for the 2013 SPE Annual Technical Conference and Exhibition, New Orleans, 30 September-2 October. The paper has not been peer reviewed. This paper presents an integrated modeling approach for history matching and economic optimization of wells producing from liquids-rich shale reservoirs (LRSRs). History matching uses daily pressures and gas-/oil-/water-production data to estimate average parameters in a 2D/3D finite-difference (FD) horizontal-multifractured-well model. Economics-based well design uses the same FD model to maximize net present value (NPV) by finding optimal well-completion parameters. Introduction Long-term historical liquid production from LRSR wells has not yet provided the industry with sufficient understanding of fundamental performance mechanisms needed to develop reliable empirical forecasting methods such as decline-curve analysis. History matching LRSR wells with a detailed FD-model description provides more-reliable liquids-production forecasting and oil-recovery predictions as well as the ability to study sensitivity of performance to model-parameter uncertainty. Combining a detailed FD model with a valid economics model provides a quantitative mechanism to optimize value by controlling well-design parameters such as horizontal-well length and fracture size and spacing. If the FD model has been history matched to an existing well, then the historical performance and economic model are “known” for the existing well, and well-design optimization can therefore study how alternative completions would have increased profitability. For new wells, optimization can be used to guide the selection of critical well-design parameters and estimate economic uncertainties. The modeling strategy presented here has been used to history match and optimize well design for LRSR wells producing from the Eagle Ford, Bakken, Avalon, and Montney formations in the US and Canada. Unfortunately, publication of field data was not permitted. For a detailed description of the model, and its use in determining bottomhole pressures (BHPs) and water-injection conditions, please see the complete paper. History Matching When history matching the FD model to observed pressure and rate data, it is preferred to control the model on BHPs measured or calculated from surface data. The most important reason to control the model on BHPs is the strong correlation between producing oil/gas ratio (OGR) and flowing BHP. Model rate performance is matched to observed data by minimizing the sum of squares (SSQ). The SSQ for the individual data types are summed together to form the total SSQ minimized through history matching. The relative size of the individual SSQ may be very different. An additional set of weighting factors, W, is used to normalize the individual SSQ. Manual, visual inspection of the model performance compared with observed data is normally required to find reasonable weighting factors. The absolute values of the unweighted SSQ are in most cases not a good indication of how well the model fits observed data. Short transients in production rates after well shut-ins are usually not captured properly by the FD model, causing large contributions to the SSQ. The weighting factors for these periods should in most cases be set to zero. Proper weighting of the SSQ is a trial-and-error process requiring careful study of data mismatch and engineering judgement.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 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 | 0.001 |
| 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".