Applying a deep-learning surrogate model to simulate and compare achievable oil recovery by different waterflood scenarios

Singh, Vishal; Ruwali, Nabindra; Pandey, Rakesh Kumar; Vaferi, Behzad; Wood, David A.

doi:10.1080/10916466.2023.2264902

Cited by 1 publication

(1 citation statement)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main reason for this is that there is a local concentration of plane streamlines in the water injection block, and the injected water is easy to channel along the dominant channel, resulting in premature water breakthrough in the oil well and ineffective water injection. Behzad Vaferi et al applied deep-learning alternative models to prove that the water injection pattern strongly affects the water channeling reduction and the ultimate oil recovery increase [7]. However, at present, studies on the formation law of micro-waterflood channels in real-model experiments are still insufficient, and it is difficult to take effective measures to deal with the problems in water channels [8,9].…”

Section: Introductionmentioning

confidence: 99%

Micro and Macro Flooding Mechanism and Law of a Gel Particle System in Strong Heterogeneous Reservoirs

Ye,

Wang,

et al. 2024

Gels

View full text Add to dashboard Cite

To address the issue of ineffective injection resulting from the consistent channeling of injected water through highly permeable channels in ultra-deep, high-temperature, high-salinity, and strongly heterogeneous reservoirs during the production process, a gel particle profile control agent suitable for high-temperature and high-salinity conditions was chosen. With the help of the glass etching visual microscopic model and the heterogeneous long core model, the formation mechanism of a water flooding channeling path and the distribution law of the remaining oil were explored, the microscopic profile control mechanism of the different parameters was clarified, and the profile control effect of macroscopic core displacement was analyzed. The research shows that the formation mechanism of a water flooding channeling path is dominated by the distribution law of the permeability section and the connection mode between different penetration zones. The remaining oil types after water flooding are mainly contiguous block, parallel throats, and multi-branch clusters. The profile control effect of gel particles on reservoir vertical heterogeneity is better than that of reservoir lateral heterogeneity. It was found that 10 wt% submicron particles with a median diameter of 600 nm play a good role in profiling and plugging pores of 5–20 μm. In addition, 10 wt% micron-sized particles with a median diameter of 2.63 μm mainly play a strong plugging role in the pores of 20–30 μm, and 5 wt% micron-sized particles with a median diameter of 2.63 μm mainly form a weak plugging effect on the pores of 10–20 μm. The overall profile control effect of 10 wt% submicro particles is the best, and changes in concentration parameters have a more significant effect on the profile control effect. In the macroscopic core profile control, enhanced oil recovery (EOR) can reach 16%, and the gel particles show plugging, deformation migration, and re-plugging. The research results provide theoretical guidance for tapping the potential of the remaining oil in strong heterogeneous reservoirs. To date, the gel particles have been applied in the Tahe oilfield and have produced an obvious profile control effect; the oil production has risen to the highest value of 26.4 t/d, and the comprehensive water content has fallen to the lowest percentage of 32.1%.

show abstract

Section: Introductionmentioning

confidence: 99%