Automated Identification of Optimal Deviated and Horizontal Well Targets
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract Horizontal or deviated wells provide a great way to maximize contact with the reservoir target of interest, reduce water and coning issues, allow for a larger drainage pattern, and as a result increase overall recovery. Placement of these wells has historically relied on history-matched simulation models, which require a multi-disciplinary team of people working over an extensive period of time. Moreover, in situations where static and dynamic reservoir models are unavailable, or are out of date, this approach can lead to inconclusive results in addition to being both cost and time prohibitive. In the present work, a new technology is developed to automate and streamline the process of optimal horizontal or deviated target identification in any type of reservoir and depositional environment. This technology relies on automated geologic and engineering workflows to map remaining oil and identify areas with high probability of success, advanced computational algorithms to perform an optimized global search with 3D pay tracking, statistical and machine learning techniques to assess neighborhood performance and geologic risk, and physics-based analytical and parametric models to forecast production. The algorithm can be fed multiple types of constrains including configuration constrains like length range, azimuth range, and deviation range as well as path constrains like zone-crossing, baffle-crossing, and fault surface crossing. An optimization engine allows the identification of targets that maximize a probability of success attribute, designed to reflect trends in key attributes known to influence production performance such as hydrocarbon pore volume, permeability, fracture intensity, baffle layers, spacing constrains, drainage maps, trends in WCT and GOR, fluid contacts, and so on. The identified targets are then further optimized using an interference analysis that selects the best set of non-interfering targets to maximize production. This framework has been successfully applied to several giant mature assets in the Middle East, North America and South America (with massive datasets and complexity), and in situations where static and dynamic reservoir models are unavailable, partially available, or are very out of date. In all these studies, hundreds of deviated or horizontal opportunities are initially identified. We then discuss key elements to consider during vetting to make sure the final set of identified opportunities are geologically sound, meet various validation criteria, and are feasible given the operational constrains.
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.
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 it