Recently, the team of Zhang Kai has made new progress in the field of reservoir production optimization, and the related research has been published in the Geoenergy Science and Engineering. The paper is titled " Simulating dynamics of pressure and fluid saturation at grid-scale by a deep learning-based surrogate reservoir modeling based on a fast-supply hybrid database and developing preliminary insights for future gas hydrate exploitations in China ".
Innovation: The current study uses a hybrid data source based on finite difference and the streamline data instead of the conventional solo finite-difference source. This method speeds up preparing the required database for the surrogate model and allows direct training algorithms to be applied in the training phase due to the rapid provision of data over long intervals. As a result, the introduced platform simplifies the data supply process, increases the efficiency of the model training, and greatly accelerates the use of the trained model in the practical usage phase.
Abstract: Advances in intelligent surrogate models have made them important tools in reservoir simulation. These tools are data-based in nature; Hence, their efficiency, besides the type of training system, always depends on the quality of the data source. Finite difference-based simulators usually provide the dynamic data required for these tools. Since the provision of this data is often time-consuming, in the process of designing an intelligent surrogate model, it is always necessary to use a recursive strategy in the training and simulation phase that imposes an additional error accumulation on the system, which is the inevitable essence of recursive algorithms. The current study uses a hybrid data source based on finite difference and the streamline data instead of the conventional solo finite-difference source. This method speeds up preparing the required database for the surrogate model and allows direct training algorithms to be applied in the training phase due to the rapid provision of data over long intervals. As a result, the introduced platform simplifies the data supply process, increases the efficiency of the model training, and greatly accelerates the use of the trained model in the practical usage phase. After establishment, the performance of the introduced surrogate model was validated by testing it on the model of the 10th SPE comparative solution project under various injection/production scenarios during a water flooding process. Consequently, satisfying results were achieved, and the reliability of the developed surrogate model was confirmed. Therefore, the model is expected to be of practical usage in other reservoirs as well. Moreover, preliminary insights were developed for adopting the introduced methodology of the present study for simulating the reservoir conditions in gas exploitation from natural gas hydrate reservoirs in China which are one of the main focuses of the 14th five-year development plan of the country from 2021 to 2025.
The Geoenergy Science and Engineering (formerly known as Journal of Petroleum Science and Engineering) covers the fields of petroleum and natural gas exploration, production and flow in its broadest possible sense. Topics include: reservoir engineering; reservoir simulation; rock mechanics; petrophysics; pore-level phenomena; well logging, testing and evaluation; mathematical modeling; enhanced oil and gas recovery; petroleum geology; compaction/diagenesis; petroleum economics; drilling and drilling fluids; thermodynamics and phase behavior; fluid mechanics; multi-phase flow in porous media; production engineering; formation evaluation; exploration methods; CO2 Sequestration in geological formations/sub-surface, etc. The latest impact factor of the journal is 5.168, and the average impact factor IF in the past 3 years is 3.646. This journal currently appears in the JCR Q1 or the Chinese Academy of Sciences ranking 2.
Paper link:
https://doi.org/10.1016/j.geoen.2023.211415
Citation:
Behzad Saberali, Naser Golsanami*, Kai Zhang*, Bin Gong, Mehdi Ostadhassan. Simulating dynamics of pressure and fluid saturation at grid-scale by a deep learning-based surrogate reservoir modeling based on a fast-supply hybrid database and developing preliminary insights for future gas hydrate exploitations in China[J]. Geoenergy Science and Engineering, 2023, 222:211415.
