This paper presents an efficient production optimization scheme for an oil reservoir undergoing water injection by optimizing the production rate for each well. In this approach, an adaptive version of simulated annealing (ASA) is used in two steps. The optimization variables updating in the first stage is associated with a coarse grid model. In the second step, the fine grid model is used to provide more details in final solution search. The proposed method is formulated as a constrained optimization problem defining a desired objective function and a set of existing field/facility constraints. The use of polytope in the ASA ensures the best solution in each iteration. The objective function is based on net present value (NPV). The initial oil production rates for each well come from capacity and property of each well. The coarse grid block model is generated based on average horizon permeability. The proposed optimization workflow was implemented for a field sector model. The results showed that the improved rates optimize the total oil production. The optimization of oil production rates and total water injection rate leads to increase in the total oil production from 315.616 MSm3 (our initial guess) to 440.184 MSm3, and the recovery factor is increased to 26.37%; however, the initial rates are much higher than the optimized rates. Beside this, the recovery factor of optimized production schedule with optimized total injection rate is 3.26% larger than the initial production schedule with optimized total water injection rate.
AbstractsA reasonable solution, to deal with oil field water problem, is to minimize the amount of water associated with oil production using effective completion lengths. This work presents an effective method to optimize wells’ completion lengths in an oil reservoir with a strong aquifer. The suggested technique is formulated as a constrained optimization problem that defines a NPV objective function and a set of existing field/facility constraints. An effective algorithm translates the completion lengths to connections number in the dynamic simulation model. In this approach, a genetic algorithm (GA), an adaptive version of simulated annealing (ASA) and a particle swarm optimization (PSO) hybridized with polytope technique are applied to maximize NPV. A comparison is given for their performances in a strong water-drive reservoir where the combinatorial effects of wells’ completion lengths (decision variables) should be addressed. Optimizing the lengths of completions leads to an increased production period, total oil production, retarding water breakthrough, reducing total water production, and finally increasing ultimate recovery. The results showed that total oil production by GA, ASA and PSO algorithm is increased by 11.0%, 2.40% and 2.22%, respectively, related to the initial case. Total water productions are decreased by GA, 9.82%, by ASA 2.11%, and by PSO 1.82% relative to the initial schedule. The best performance belongs to the GA algorithm. Moreover, the average watercut of all wells is decreased through the optimization process. Besides, based on the numerical simulation, closing the worst connections with high watercut decreases total water production, and improves oil recovery, maximum well productivity, and NPV (oil–water ratio is increased 18.2%). Most connections are placed in the layers where water coning can occur later (considering near-well-bore permeability) and slightly far from full water zone.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.