Physics-Informed Neural Networks (PINNs) gains attentions as a promising approach for applying deep neural networks to the numerical solution of nonlinear partial differential equations (PDEs). However, due to the challenging regions within the solutions of 'stiff' PDEs, e.g., shock front of CO₂ immiscible flooding, adaptive methods are essential to ensure the neural network accurately addresses these issues. In this work, we introduce a novel method for adaptively training PINNs, named Self-Adaptive PINNs (SA-PINNs). This approach employs fully trainable adaptation weights that are applied individually to each training point. Consequently, the neural network autonomously identifies challenging regions of the solution space and focuses its learning efforts on these areas. This method is hereby used to simulate a two-phase immiscible flooding in a low-permeability oil reservoir, with considering gas dissolution and the threshold pressure gradient of oil phase in low-permeability oil reservoirs, i.e., modified Buckley-Leverett (B-L) problem. The model is capable of generating a precise physical solution, accurately capturing both shock and rarefaction waves under the specified initial and boundary conditions, though the introduction of complicated physics increases the nonlinearity of the governing PDEs. The self-adaptive mechanism modifies the behavior of the deep neural network by simultaneously minimizing the losses and maximizing the weights. It, thus, can effectively capture the non-linear characteristics of the solution, thereby overcoming the existing limitations of PINNs. In these numerical experiments, the SA-PINNs demonstrated superior performance compared to other state-of-the-art PINN algorithms in terms of L2 error. Moreover, it was also achieved with a reduced number of training epochs. SA-PINNs can effectively model the dynamics of complex physical systems by optimizing network parameters to minimize the residuals of the PDEs.

@en

Steam-assisted gravity drainage (SAGD) is one efficient and mature technology for recovering heavy oil and bitumen resources. The key underlying mechanism is the growth of the steam chamber after injecting steam. However, due to the complex geological environment, the thief zones exist and have a prejudicial effect on the development of the steam chamber, thus impacting the ultimate heavy-oil recovery. In this work, our objective is to investigate the effect of a top-water thief zone (i.e., water zone overlies the oil sand) on SAGD performance and further to understand the crucial mechanisms that control the heat loss during steam injection. A large-scale three-dimensional experimental apparatus is used to carry out the SAGD process with a top aquifer. Based on the similarity criterion, the field-scale model is transformed into a laboratory elemental model. To evaluate the SAGD performance quantitatively, the dynamic growth of the steam chamber is measured using the thermal detectors and the production data is recorded. The results show that the steam chamber exhibits three distinguished stages, that is, upward spread, lateral extension, and downward development in the presence of top-water zone. The bottom-water zone has less impact on the steam-chamber growth. The existence of a confined top-water zone, however, significantly affects SAGD performance, especially the lateral expansion of the steam chamber. The lateral propagation of the steam front is hindered by the top thief zone due to the heat exchange with the top water. Once the steam chamber reaches the boundary, the accumulation of energy in the water thief zone, in turn, can reduce the remaining oil saturation along the topwater-oil interface. This study provides us some key insights into the development of heavy oil resources with top thief zones when implementing SAGD technology.@en

Aromatic micropollutants found in wastewater pose threats to human health, and it is necessary to remove them from wastewater. Cyclodextrin hydrogel is one of the most promising environmentally friendly materials for removing aromatic micropollutants from wastewater by means of adsorption. In this study, we have synthesized a cyclodextrin hydrogel designed to simultaneously adsorb a cationic dye (methylene blue, MB) and an endocrine-disrupting compound, EDC (bisphenol A, BPA) and provide insights into the design of adsorbents for multicomponent wastewater purification. Polymeric dextran-poly(acrylic acid)-β-cyclodextrin (Dex-PAA-CD) was prepared by the copolymerization of glycidyl methacrylate (GMA), methacrylate-substituted dextran (Dex-MA), acrylic acid, and vinyl-derivatized β-cyclodextrin (β-CD-MA). The adsorption of MB and BPA by the Dex-PAA-CD hydrogel presented short equilibrium time. The adsorption of MB and BPA was independent of each other, and their adsorption mechanisms were proved to be electronic interaction and cyclodextrin host-guest interaction, respectively. The maximum adsorption capacities of MB and BPA onto Dex-PAA-CD are determined to be 5.60 and 0.144 mmol/g by the Sips isotherm. The Dex-PAA-CD hydrogel retains a high adsorption capacity even after five adsorption/desorption cycles and demonstrates the feasibility of designing reusable adsorbents to remove multiple pollutants from wastewater.

@en

Steam channeling, one serious problem in the process of steam flooding in heavy oil reservoir, decreases the sweep efficiency of steam to cause a lower oil recovery. Viscosity reducer and nitrogen foam, two effective methods to improve oil recovery with different mechanism, present a satisfactory result after steam flooding. In this article, a 2D visualized device was introduced to investigate the synergistic development effect of two different chemical additives and intuitively study their flowing characteristic in porous media, as well as macroscopic and microscopic mechanism of improving heavy oil recovery by chemical additives after steam flooding. The results showed that the fingering phenomenon was generated obviously in the process of steam flooding, which restricted the swept area of steam. Due to decreasing oil-water interface tension, O/W emulsion with lower viscosity was formed to enhance the oil flow capacity and polish up the displacement efficiency of steam after injecting viscosity reducer. And the synergistic effect of viscosity reducer & foaming agent was more conductive to improve displacement efficiency of steam, with 4.3% of oil recovery higher than purely viscosity reducer assisting steam flooding in this process. Microscopic results indicated that thermal foams can be trapped in the porous media to improve injection profile effectively and displace the residual oil caused by steam flooding. The ultimate oil recovery of synergistic development is 65.6%, 11.0% higher than one additive (viscosity reducer). This article can provide reference for the study of thermochemistry assisted steam flooding in heavy oil reservoir.

@en