A Guided Pattern Search with a non-intrusive reduced order modeling for oil production optimization: Brugge field case study

An efficient optimization process for hydrocarbon production in presence of geological uncertainty using a clustering method: A case study on Brugge field

“…In fact we introduce uncertain parameters as input variables of ANN when we define and train it. For more details about the methodology and formulation of deterministic method refer to Foroud et al (2016).…”

“…Recently, GPS has been introduced and examined in case of deterministic optimization of oil production which led to promising results (Foroud et al, 2016). In this paper the merits of this method will be investigated in case of production optimization in the presence of geological uncertainties.…”

“…Using one of these matched realizations, a deterministic optimization of the production plan in this field has been performed before (Foroud et al, 2016). Here, we are going to involve multiple geological realizations (40 matched models) in the well control optimization.…”

“…The well controls include both injection and production flow rates and the objective function is to maximize discounted NPV. Besides the earth model selection, we also consider the use of previously developed ROM and a GPS algorithm for production optimization in presence of uncertainty (Foroud et al, 2016). Applying a ROM technique based on Discrete Empirical Interpolation Method (DEIM) coupled with an Artificial Neural Network (ANN) will provide significant reduction in optimization run time by decreasing computational efforts involved in reservoir simulation.…”