Robust waterflood optimization under geological uncertainties using streamline-based well pair efficiencies and assimilated models

Abstract

Effective management of waterflooding requires optimization of injection and production rates. However, nominal optimization methods based on a single realization can result in suboptimal hydrocarbon recovery and sweep efficiency due to reservoir heterogeneity. To account for geological uncertainties, this paper proposes a streamline-based robust rate optimization that iteratively adjusts well rates of each realization using a joint well pair rate multiplier calculated from optimal interpolation method. This optimization process is applied over time intervals until the end of field life. The effective and robustness of this approach are validated using a 2D synthetic case and then applied to Brugge benchmark case. For each application, we generate 100 history-matched geological realizations, evenly split into 50 for training and 50 for testing purposes. Robust optimization is employed on the complete training dataset to derive an optimal schedule that maximizes the expected oil across all realizations. Simultaneously, nominal optimization is individually applied to each training realization to establish an optimal schedule specific to each instance. Subsequently, both the robust optimization schedule and the nominal optimization schedules are applied to the testing dataset to assess and compare their effectiveness. Results indicate that the streamline-based rate allocation optimization effectively adjusts well rates based on well-pair efficiency. The robust optimized schedule takes into account geological uncertainties and enhances cumulative oil production, even if it doesn't consistently outperform the nominal optimized schedules. Overall, this approach provides an effective way to optimize waterflooding management under geological uncertainties.