The main objective of modern reservoir management is to maximize the oil recovery when a displacing agent, such as CO2, is injected to displace the residual oil in a reservoir. Such process can be controlled properly by allocating the injected fluids to the injectors and adjusting the produced fluids from the producers. Inappropriate production-injection strategy leads to early breakthrough, unstable pressure distribution, and low ultimate oil recovery. Furthermore, presence of physical and/or financial uncertainties elevates the complexity of the field production optimization. In this paper, a pragmatic technique has been developed and successfully applied to determine the optimum production-injection strategy in a CO2 flooding reservoir by incorporating well performance into reservoir simulation in the presence of both physical and financial uncertainties. More specifically, well rates of the injectors and flowing bottomhole pressures of the producers are chosen as the controlling variables. Several variable candidates are first assessed, determined and finally assigned to each well based on the inflow performance curve, multiphase flow behavior in the wellbore, and voidage balance within the reservoir. An objective function associated with both the average net present value (NPV) and the range of NPV uncertainty is then defined, while a modified genetic algorithm is utilized as optimization engine to determine the optimum production-injection strategy. In addition, multiple equal-probable reservoir models are used to account for the physical uncertainty, while prices of oil and CO2 are applied to assess and quantify the financial uncertainty. Compared to the production-injection strategies without optimization, it is shown from a field case that the optimum strategy can postpone the CO2 breakthrough time by 1.5 years, decrease the water cut by 8.4%, and increase the oil recovery and net present value by 7.8% and 6.6%, respectively.

You can access this article if you purchase or spend a download.