Key Technologies Study On Carbon Dioxide Flooding In Depleted Reservoir

With the rapid development of national economy in recent years, the consumption of oil is more and more serious in our life. Meanwhile, the greenhouse effect seriously affects the life of people, which is caused by the carbon dioxide emission from the increasing fossil energy consumption. The depleted reservoir is selected for Carbon dioxide to inject in this paper for which can provide favorable storage place, and can use carbon dioxide to improve its recovery factor.According to the characteristics of the depleted reservoir, a random porosity distribution model with characteristic of normal distribution is used in the study to characterize the reservoir heterogeneity. Because most pores of the reservoir is occupied by water after water flooding, residual oil is dispersed in the corner or small pores where water flooding can’t reach.(1) Coupled with the viscosity and density equation, two phase flow mathematical model is established between CO2 and the formation water. Results show that the saturation distribution and the oil displacement effect are impacted greatly by the porosity heterogeneity of formation which has little impact on the injection point pressure. But the injection point pressure increases gradually with the depth of reservoir, and it also increases gradually with the injection time until stabilized eventually at one time point. The phenomenon of gravity overlap exists during CO2 flooding due to the density difference between two fluid. Under the same storage space, the amount of carbon dioxide storaged in deep oil reservoir is relatively smaller than in the shallow one. Even the deep reservoir required higher injection pressure, it also has great advantage for its seal effect.(2) Rock deformation, energy transmission, fluid flow and solute transport process are considered when CO2 contacting the remaining oil and rock. And comsolmultiphysics 3.5 a is used to simulate the process. The simulation results show that:After the same displacement time, CO2 flooding area decreases gradually as rock stress increases, but increases with the rise of temperature and injection pressure. The injection pressure plays a main role on oil displacement effect in these factors. And the viscous coefficient of crude oil decreases gradually during the whole driving time.(3) Carbon dioxide gravity steady drainage was considered and viscous fingering model was established in the depleted reservoir which has high closure, large stratigraphic dip, thick reservoir and high vertical permeability. The results indicate that CO2 has a good viscosity reduction property for heavy oil. The numerical simulation was implemented under several geological and operating conditions including positive rhythm reservoir, reverse rhythm formation, permeability, viscosity and velocity of displacement.(4) Artificial neural network was used to forecast the CO2 flooding recovery efficiency. The prediction results show that the output value of the neural network forecasting has 8% error rate contrast with the expectation, and the correlation coefficient between the curve of the prediction results and mathematical model is 0.93, which has a high accuracy. Therefore, using artificial neural network to predict the recovery efficiency of carbon dioxide flooding is feasible, which can be deemed to be an effective method in predicting CO2 flooding oil recovery.