The use of injection and production rate control is a common practice for optimal waterflood management. However, the benefits of rate control for enhanced oil recovery (EOR) processes have not been fully explored. In this paper, we examine the role of rate optimization in polymerflooding to maximize sweep efficiency and to minimize polymer recycling by delaying polymer breakthrough.

Field scale rate optimization problems for EOR processes involve complex physical and simulation models, production and facility constraints, and a large number of unknowns. Deployment of smart well completions with inflow control valves (ICV) to control production/injection rates for various segments along the wellbore further compounds to the complexity of the optimization. We propose a practical and efficient approach for computing optimal production/ injection rates for polymerflooding with application to smart wells. Our approach relies on equalizing arrival time of the floodfront at all producers to maximize the sweep efficiency and additional ‘norm’ constraints on the arrival times to achieve production acceleration. The ‘optimal’ rate strategy is decided based upon a compromise between maximizing sweep efficiency and production acceleration. We use streamlines to compute the analytical sensitivity of arrival times with respect to well rates. Analytical forms for gradient and Hessian of the objective function are derived, making our optimization computationally efficient for large-scale applications under a hierarchy of operational and facility constraints.

We demonstrate our approach using a 2D synthetic example and a field-scale application. The results clearly demonstrate the benefits of rate optimization either by reducing polymer usage or increasing oil recovery. In particular, the 3D field-scale application results in achieving higher oil recovery and reducing associated water and polymer production for the same amount of polymer injection. The geological uncertainty has been accounted for via a stochastic optimization framework based on a combination of the expected value and variance of a performance measure from multiple realizations.

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