A new color‐gradient lattice Boltzmann model is developed to simulate spontaneous imbibition in a porous media micromodel, which needs only two‐dimensional computational cost but considers essential three‐dimensional effects. A modified periodic boundary condition is introduced to deal with inlet and outlet boundaries with great ease and effectiveness. This model is first validated against analytical solutions and micromodel experiment. It is then used to study the spontaneous imbibition process in a homogeneous micromodel for varying viscosity ratios and contact angles. Depending on the viscosity ratio (λ) of wetting to nonwetting fluids, three different imbibition patterns are observed, namely unstable displacement (UD), stable displacement (SD), and crossover from UD to SD, which occur at , , and , respectively. When the crossover occurs, the wetting fluid saturation at breakthrough increases between two plateaus corresponding to UD and SD, respectively. Consistent with theoretical predictions from a single capillary, the wetting fluid saturation versus time follows the relation of for , and for . Due to the increased capillary valve resistance, increasing wetting phase contact angle is found to promote the flow instability, resulting in the development of preferential flow paths and thus some nonwetting fluid trapped in the flow direction. In addition, it is theoretically and numerically demonstrated that the effect of contact angle cannot be properly described by the theoretical equations derived from a single capillary because of the presence of capillary valve effect, which can be suppressed by decreasing contact angle or micromodel depth.

The closed-loop reservoir management technique enables a dynamic and real-time optimal production schedule under the existing reservoir conditions to be achieved by adjusting the injection and production strategies. This is one of the most effective ways to exploit limited oil reserves more economically and efficiently. There are two steps in closed-loop reservoir management: automatic history matching and reservoir production optimization. Both of the steps are large-scale complicated optimization problems. This paper gives a general review of the two basic techniques in closed-loop reservoir management; summarizes the applications of gradient-based algorithms, gradient-free algorithms, and artificial intelligence algorithms; analyzes the characteristics and application conditions of these optimization methods; and finally discusses the emphases and directions of future research on both automatic history matching and reservoir production optimization.

The permeability of sediments is an important parameter that effects flow characteristics of the gas and water in the process of natural gas hydrates development. The distribution of gas hydrate is generally heterogeneous in porous media. It poses a great challenge to the permeability measurement of hydrate‐bearing sediments. To study the effect of heterogeneous distribution of methane hydrate on the permeability of porous media, methane hydrate formation in the sandstone and permeability measurement was carried out using a low‐field nuclear magnetic resonance (NMR) measurement system in situ conditions. The spatial distribution of methane hydrate in the core samples was determined by combining the T₂ distribution with magnetic resonance imaging. The results show that the distribution of methane hydrate is heterogeneous in the core samples. The temporal and spatial variation of permeability happens in the process of hydrate dissociation. The heterogeneity of methane hydrate distribution severely affects the analysis of permeability change. Considering the heterogeneity of methane hydrate distribution, a new method was proposed to obtain the quantitative relationship between hydrate saturation and relative permeability based on magnetic resonance imaging and the equivalent seepage capacity method. The results show that the relative permeability models obtained by this method agree better with experimental results. The optimum coefficients of relative permeability models obtained by equivalent seepage capacity method change significantly in contrast with the direct fitting method. The variation of the optimum coefficient is up to 300% in this study. The model coefficients are related to the pore characteristic of hydrate‐free porous media.