Data Sufficiency for Reservoir Development Decision-Making in the Presence of Uncertainty
Bibliographic record
Abstract
Abstract Uncertainty stemming from the sparse information available to model a reservoir and from the lack of complete knowledge about the flow processes in the reservoir is an inescapable aspect of reservoir modelling. Methodologies for assessing and incorporating that uncertainty in exploration and development decisions are crucial for successful management of assets. This paper presents a reservoir modelling case study demonstrating typical steps in the exploration of a reservoir and aspects of uncertainty assessment using geostatistical techniques. The viability of using simple, easy-to-use connectivity measures for assessing the productivity of candidate well locations prior to actually drilling the well is explored. A Bayesian method for incorporating seismic data in reservoir models is presented. Introduction The true distribution of facies, porosity, and permeability in a petroleum reservoir cannot be uniquely determined using the information from a few widely spaced wells. Geological uncertainty in the form of local uncertainty (the uncertainty in an attribute value at any given location in the reservoir) and joint uncertainty (uncertainty in the connectivity characteristics of the reservoir) are an inescapable aspect of reservoir modelling. Decisions such as collection of additional data for improved reservoir delineation, location of additional wells for reserves depletion, and implementation of improved oil recovery would have to be based on an assessment of geological and reservoir response uncertainty and the potential impact of additional data on that ncertainty. Geostatistics provides a framework for modelling reservoirs taking into account data from diverse sources such as well logs, geological outcrops, seismic, and well/reservoir production information. The spatial distribution of reservoir attributes is modelled as the manifestation of a spatial random function. Under this random function hypothesis, the reservoir attribute at every location within the reservoir is modelled as a random variable (RV). The probability distribution characterizing the RV represents the uncertainty in the reservoir attribute value at that location. The local conditional probability distribution reflecting the uncertaintyin attribute value at a particular location is constructed using a interpolation technique such as kriging. The random function is characterized by a multivariate, joint probability distribution corresponding to the distribution of RV at all locations within the reservoir taken jointly. Stochastic models of the reservoir are btained by sampling realizations from this joint multivariate distribution. Reservoir modelling using geostatistics therefore consists of generating multiple realizations of the reservoir that reflect the uncertainty due to incomplete information. This paper presents a methodology for integrating seismic and well data in order to develop stochastic representations of the reservoir. Decisions pertaining to reservoir development in the presence of uncertainty are evaluated. Since geostatistics provides a framework for integrating data from diverse sources and for assessing geologic uncertainty, the issue of data sufficiency and the worth of additional reservoir specific information will be examined using geostatistical tools. The modelling methodology nd techniques for assessment of uncertainty are demonstrated on the Stanford V reservoir(1), a synthetic data set developed specifically to test the accuracy of reservoir characterization techniques.
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 | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 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".