Gravity can reduce the instability of the gas-oil contact that is caused by gas channeling in locations with low flow resistance, such as high-permeability layers, macropores, and fractures during the gas-assisted gravity drainage process. Herein, the microscopic forces during the gas-assisted gravity drainage process were analyzed and combined with the capillary model to study the occurrence boundary of gas-assisted gravity drainage process, and the characteristics of the gas-oil contact in the gas-assisted gravity drainage process was discussed. The results show that free gravity drainage occurs only in pores where a certain height of the oil column and pore radius are reached. Furthermore, the lower the oil-gas interface migration rate, the easier free gravity drainage occurs. In other scenarios, additional gas injection is required. During the gas-assisted gravity drainage process, the gas-oil contact moves down stably as a transition. The width of the transition zone and the available pore radius are related to the gas-oil contact migration rate and the oil viscosity; the smaller the gas-oil contact migration rate and the lower the oil viscosity, the smaller pore throat can be involved in mobilization. Optimizing the gas injection rate and reducing the oil viscosity can delay the gas channeling maturity time, which is beneficial for the realization of the gas-assisted gravity drainage process. Finally, a method considering micropore heterogeneity is established for determining the critical gas injection rate, while the mainstream pore throat can be involved in mobilization and the gas-oil contact can be stabilized at the same time. The method of determining the critical gas injection rate can help researchers and reservoir engineers to better understand and implement the gas-assisted gravity drainage process.

A compositional reservoir simulation model with ten-million grids is successfully computed using parallel processing techniques. The load balance optimization principle for parallel calculation is developed, which improves the calculation speed and accuracy, and provides a reliable basis for the design of reservoir development plan. Taking M reservoir as an example, the parallel numerical simulation study of compositional model with ten million grids is carried out. When the number of computational nodes increases, message passing processes and data exchange take much time, the proportion time of solving equation is reduced. When the CPU number increases, the creation of Jacobian matrix process has the higher acceleration ratio, and the acceleration ratio of I/O process become lower. Therefore, the I/O process is the key to improve the acceleration ratio. Finally, we study the use of GPU and CPU parallel acceleration technology to increase the calculation speed. The results show that the technology is 2.4 $\sim$5.4 times faster than CPU parallel technology. The more grids there are, the better GPU acceleration effect it has. The technology of parallel numerical simulation for compositional model with ten-million grids presented in this paper has provided the foundation for fine simulation of complex reservoirs.

The crosslinked polymer flooding, which is developed on the basis of polymer flooding, is a new type of flooding technology. As an EOR method, cross-linked polymer flooding has become a research hotspot. In the process of cross-linked polymer flooding, if the concentrations of the polymer and the crosslinking agent are small, the viscosity of the solution is low, and it will not achieve the oil displacement effect. Meanwhile, if the concentrations of the polymer and crosslinking agent are large, the viscosity of the solution is high, it needs high pressure to drive it flowing in the formation. Further, with the increasing injection of chemical agents, the contradiction between reduced production and increased cost has presented. The performance of crosslinked polymer flooding depends on the interaction of these two factors. Therefore, the concentrations of polymer and crosslinking agent should be optimized. In this paper, an optimal design method is proposed by using genetic algorithm with global optimization characteristics algorithm, combining with the chemical flooding numerical simulation software UTCHEM, the concentrations of the chemical agents are optimized. Firstly, the cumulative oil production is calculated by numerical simulation software UTCHEM, then the concentrations of the chemical agents are randomly generated by the genetic algorithm in the encoding process, and the fitness function takes the profit of cross-linked polymer flooding. Given a set of initial values, through crossover and mutation of population, optimized injection concentrations of the polymer and cross-linking agent are obtained by the multi-generational calculation.