Multiple-Point Simulation of Multiple Reservoir Facies
Abstract/Contents
- Abstract
- Multiple-point (mp) statistics has taken roots in geostatistics. Training images, as source of structural information, are at the core of multiple-point statistics. These training images are the equivalents of variogram models used in traditional 2-point geostatistics. Training images are more general than variogram models in that mp in addition to 2-point statistics can be extracted from them. However, multiple-point statistics can also be used to capture the high entropy patterns of training images generated from Gaussian field fully characterized by 2-point statistics. As long as the training image depicts the correct Gaussian-type structures, the mp simulation algorithm would generate results as good as those obtained from direct sequential Gaussian simulation.An improved algorithm to retrieve from training images conditional probability distribution functions (cpdf) is proposed. The cpdf is built by weighting all data events present in the search template. Compared with the original algorithm, which retained only the single largest data event with enough replicates, the proposed algorithm significantly improve reproduction of the large scale structures present over the training image.As assumed in 2-point geostatistics, stationarity is an indispensable prior decision allowing the inference of multiple-point geostatistics. A training image must reflect stationary patterns, i.e. patterns which are repeated often enough over the training image so that they can be captured by multiple-point statistics.Actual reservoirs typically contain nonstationary structures/patterns which are specific to some locations or areas. Location-dependent linear transforms (rotation + affinity) can be used to transfer the stationary patterns of a training image to locations of the specific reservoir being modeled, resulting in simulated patterns which are nonstationary and location-dependent. This approach is similar to that of building complex and locally varying variogram models from a set of basic stationary and isotropic variogram models. Such basic stationary training images are called training image modules.The flowcharts of newly developed snesim program 5.0 written by Strebelle and corresponding remarks are given in the last section.
Description
| Type of resource | text |
|---|---|
| Date created | May 2002 |
Creators/Contributors
| Author | Zhang, Tuanfeng |
|---|---|
| Primary advisor | Journel, Andre G. |
| Degree granting institution | Stanford University, Department of Petroleum Engineering |
Subjects
| Subject | School of Earth Energy & Environmental Sciences |
|---|---|
| Genre | Thesis |
Bibliographic information
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
Preferred citation
- Preferred Citation
- Zhang, Tuanfeng. (2002). Multiple-Point Simulation of Multiple Reservoir Facies. Stanford Digital Repository. Available at: https://purl.stanford.edu/yn566yz1577
Collection
Master's Theses, Doerr School of Sustainability
View other items in this collection in SearchWorksContact information
- Contact
- brannerlibrary@stanford.edu
Also listed in
Loading usage metrics...