Real-Time Optimization for Reservoir Management with Conventional and Intelligent Wells

Abstract

The use of a real-time optimization (OTR) of an oil field can increase oil recovery, reduce water production, mitigate the risks and avoid intervention in wells, maximizing profits. However, a clear methodology that shows advantages of optimizing stopped in many times, due to complexity of the problem because the operation of many conventional or intelligent wells involves a large number of control variables, which makes the optimization process spend a high computational time, often making unfeasible a more complete approach. This complexity increases with the number of wells and valves used in the simulation model, which will be used to estimate the possible gains. To circumvent this problem, this project proposes the development of an optimization method that can be employed to perform OTR, improving the management of the field. The method proposed involves using a genetic algorithm initially then to make it faster and more efficient by reducing the solution space, leading to a lower computational time in the simulations during optimization process. Another approach is to obtain an expression that relates the flow rate of the well (or valve aperture) with the WCUT (water cut) and RGO (ratio of gas-oil) of the well, accelerating the optimization process. Once achieved greater efficiency and speed in the optimization process, a complex reservoir model can be used to apply the method, and can fill an important gap in the literature of OTR, which means that many companies fail to evaluate clearly the real benefits and advantages of the operation of the wells in optimal conditions.