Application of Machine Learning Technologies for Rapid 3D Modelling of Inflow to the Well in the Development System
Abstract
The most convenient and physically justified tool for optimization of oil field development system is 3D hydrodynamical modeling. The high resource consumption of the method, however, does not allow carrying out fast multivariate calculations. In this paper, we propose an approach to 3D modeling in a development unit, based on the machine learning algorithms. According to available data of the reservoir properties and using high-performance computing systems, we've created a training set and a metamodel of the development unit. Forecasting in the latent variables space makes it possible to perform the necessary calculations many times faster than for the classical approach. We also offer a method for improving the quality of calculations on coarsened grids, that allows to greatly increase the size of training set and to reduce the complexity of setting up a metamodel.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Modern surrogate reservoir models with a new computation engine based on deep neural networks promise 100-1000 times the conventional models' speedup while keeping similar functionality.
“Modern surrogate reservoir models with a new computation engine based on deep neural networks compress the mathematical problem dimensionality and approximate the time derivatives promise 100-1000 times the conventional models' speedup while keeping similar functionality [ 28 , 29 ].”
From Artificial intelligence in oil and gas upstream: Trends, challenges, and scenarios for the future (2021) · cites this paper as [29]
Cites
Efficient Brownfield Optimization of a Reservoir in West Siberia
Ushmaev, O. S., Babin, V. V., Glavnov, N., et al. (2017) · Proceedings · doi:10.3997/2214-4609.201700259 · Cited by 5 · see citing works
Grid Cluster in the Office: High-Performance Computing for Reservoir Management
Yaubatyrov, R., Babin, V. V., & Akmadieva, L. (2018) · SPE Annual Technical Conference and Exhibition · doi:10.2118/191519-ms · Cited by 5 · see citing works
Flagged first, then by citations
- Authors
- Simonov, M., Akhmetov, A., Temirchev, P., Koroteev, D., Kostoev, R., Burnaev, E., & Oseledets, I.
- Affiliations
- LLC, Gazpromneft Science & Technology Centre
- Skolkovo Institute of Science and Technology (RU)
- Venue
- SPE Russian Petroleum Technology Conference
- Publisher
- SPE
- Published
- 2018-10-15
- OpenAlex
- W4237924618
- Cited by
- 5
- References
- 2
- Citations by year
- 2026: 1 · 2024: 1 · 2020: 3
- FWCI
- 1.46
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- SPE Russian Petroleum Technology Conference (publishedVersion)
- Topics
- Physical Sciences › Engineering › Ocean Engineering › Reservoir Engineering and Simulation Methods
- Physical Sciences › Engineering › Ocean Engineering › Enhanced Oil Recovery Techniques
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Model Reduction and Neural Networks
- Keywords
- machine learning, 3D reservoir modeling, hydrodynamic modeling, meta-model, latent variables, training data, high-performance computing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
