Bibliographic record
Abstract
In the 2020 Completions Technology Focus, I stated that digitization will forever change how the most complex problems in our industry are solved. And, despite another severe downturn in the upstream industry, data science continues to provide solutions for complex unconventional well problems. Casing Damage Casing collapse is an ongoing problem and almost always occurs in the heel of the well. It prevents passage of frac plugs and milling tools. Forcing a frac plug through the collapsed section damages the plug, predisposing it to failure, which leads to more casing damage and poor stimulation. One team has developed a machine-learning (ML) model showing a positive correlation between zones with high fracturing gradients and collapsed casing. The objective is a predictive tool that enables a completion design that avoids these zones. Fracture-Driven Interactions (FDIs) Can Be Avoided in Real Time Pressurized fracturing fluids from one well can communicate with fractures in a nearby well or can intersect that well-bore. Such FDIs can occur while fracturing a child well and can negatively affect production in the parent well. FDIs are caused by well spacing, depletion, or completion design but, until recently, were not quickly diagnosed. Analytics and machine learning now are being used to analyze streaming data sets during a frac job to detect FDIs. A recently piloted detection system alerts the operator in real time, which enables avoidance of FDIs on the fly. Data Science Provides the Tools Analyzing casing damage and FDIs is a complex task involving large amounts of data already available or easily acquired. Tools such as ML perform the data analysis and enable decision making. Data science is enabling the unconventional “onion” to be peeled many layers at a time. Recommended additional reading at OnePetro: www.onepetro.org. SPE 199967 - Artificial Intelligence for Real-Time Monitoring of Fracture-Driven Interactions and Simultaneous Completion Optimization by Hayley Stephenson, Baker Hughes, et al. SPE 201615 - Novel Completion Design To Bypass Damage and Increase Reservoir Contact: A Middle Magdalena, Central Colombian Case History by Rosana Polo, Oxy, et al. SPE 202966 - Well Completion Optimization in Canada Tight Gas Fields Using Ensemble Machine Learning by Lulu Liao, Sinopec, et al.
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 | 0.001 | 0.001 |
| 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".