Statistical Pattern Recognition and Geostatistical Data Integration
Bibliographic record
Abstract
SummaryStatistical pattern recognition, particularly the neural network approach, has found many applications in reservoir characterization, enabling the use of multi-variate, imprecise and uncertain reservoir data. Geostatistics is a well-established field for 3D spatial modeling and uncertainty quantification of the reservoir facies and petro-physical properties. In this paper we present a theoretical and practical framework for developing and applying pattern recognition tools within the traditional geostatistical framework. We show that the power of soft computing tools within a geostatistical framework allows the modeler to make maximum use of the reservoir data. Geostatistics aims at integrating geophysical and reservoir engineering data, yet at the same time honoring geological continuity information provided by well data or by analog outcrop information. However, the traditional geostatistical framework does not allow an easy integration of “non-linear” reservoir data. Due to the nature of the governing physical laws, seismic amplitude data and production history data both exibit a non-linear and multiple point relationship with petrophysical properties such as porosity and permeability. In this paper we show how statistical pattern recognition tools can be integrated into traditional and novel geostastical simulation methods in order to deal with the imprecise and non-linear aspects of reservoir data. Probabilistic type neural networks such as the proposed logistic regression network are ideal tools to model the probabilistic relation between reservoir data and reservoir properties. The output of these types of neural networks is a conditional probability, rather than the single estimate provided by more traditional neural networks. A framework is presented where these networks can be integrated within any geostatistical simulation algorithm. We provide two examples of this novel approach. First we show how neural networks axe trained to build a non-linear relation between seismic amplitude data and reservoir facies. The trained neural network is used to constrain a fluvial reservoir to seismic amplitude data. A second example shows the worth of using neural networks in understanding and calibrating the non-linear relationship between the permeability heterogeneity and well test response data. The resultant neural network calibrated relationship is then used to condition multiple reservoir models to well test data using an iterative Gaussian simulation method.
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 | 0.003 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
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".