Hybrid mathematical modelling of three-phase flow in porous media: Application to water-alternating-gas injection
نویسندگان
چکیده
Machine learning algorithms are extensively used to reduce the complexity of applied problems in various fields, including energy. Accurate prediction performance water alternating gas (WAG) injection as an enhanced oil recovery (EOR) process is great importance optimal management hydrocarbon resources. In current work, a hybrid mathematical model proposed for near-immiscible WAG process. We use data-driven sub-models, least square support vector machine (LSSVM) and adaptive neuro-fuzzy inference system (ANFIS) series with empirical (EM) first principle (FPM) study three-phase flow porous media. The LSSVM ANFIS sub-models predict two-phase water-oil, gas-oil, gas-water relative permeabilities. outputs from these models supplied (EMs) estimate permeabilities oil, gas, phases. developed using shows better estimating permeabilities, compared that ANFIS. parameter analysis reveals sub-model, saturation most influencing input oil-water systems while gas-oil system, important parameter. Using this some forecast ultimate factor (RF) testing phase. predicted RF values 92.0%, 91.6%, 82.9% correlation-based EM-FPM, LSSVM-EM-FPM, ANFIS-EM-FPM models, respectively, comparison measured value 93.6% after three cycles gasinjection. Among LSSVM-EM-FPM significantly removes non-linearity general, possesses same level accuracy EM-FPM model, but less non-linearity. Thus, can be demanding applications such optimization control process, leading resource management.
منابع مشابه
Mathematical Models of Dusty Gas Flow through Porous Media
This work reports on the recent advances in the continuum approach to dusty gas flow modeling through isotropic porous structures. This approach has received considerable attention over the last half century due to the need to develop dusty gas flow models capable of describing natural and industrial transport phenomena, including subsurface transport of dissolved or suspended particulates, des...
متن کاملMathematical Modelling of Dusty Gas Flow through Isotropic Porous Media with Forchheimer Effects
Dusty gas flow through isotropic porous media is considered. The equations governing dusty gas flow through free space are intrinsically averaged in order to derive a comprehensive model that describes flow of a dusty gas through porous media. The developed model is capable of describing flow of mixtures with non-uniform number density through variable porosity media, while taking into account ...
متن کاملGas-Water Flow Behaviour During Mutual Displacement in Porous Media
During oilfield development, there exist multi-cycle gas–water mutual displacement processes. This means that a cycling process such as water driving gas–gas driving water–water driving gas is used for the operation of injection and production in a single well (such as foam huff and puff in single well or water-bearing gas storage). In this paper, by using coreand micro-pore scales model, we st...
متن کاملA Novel Approach to Measuring Water and Oil Relative Permeabilities in Two-phase Fluid Flow in Porous Media
In this study, direct laboratory measurements of unsteady-state imbibition test are used in a new approach to obtain relative permeability curves with no predetermined functionality assumptions. Four equations of continuity, Darcy’s law, cumulative oil production, and water fractional flow are employed in combination together under certain assumptions to present the new approach which interpret...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of Natural Gas Science and Engineering
سال: 2021
ISSN: ['2212-3865', '1875-5100']
DOI: https://doi.org/10.1016/j.jngse.2021.103966