Closed-loop stochastic oilfield optimization for hedging against geologic, development, and operation uncertainty

Abstract

Development and management of oilfields involve several sources of uncertainty that complicate an already challenging decision-making process. Two main sources of uncertainty are related to geologic description of reservoirs and future development and operation scenarios. While geologic uncertainty has been widely studied and robust optimization methods have been developed to account for it, the uncertainty in future development and operations has not been considered. Traditionally, future development strategies have been included as optimization decision variables. However, in practice, due to unpredictable factors, future plans tend to deviate from the solutions obtained in optimization problems. In a recent publication, we showed that a more prudent and realistic approach toward oilfield optimization is to consider future development plans as uncertain parameters to ensure that the obtained solutions remain robust when future plans change. Here, we develop a closed-loop stochastic field development optimization formulation to account for the uncertainty in geologic description and future development and operation scenarios. The proposed approach optimizes the decision variables for current stage of planning (e.g., well locations and operational settings) while accounting for geologic and future development and operation uncertainties. Geologic uncertainty is represented with several reservoir model realizations, while development uncertainty is incorporated through drilling scenario trees and stochastic well placement. To account for operation uncertainty, future well controls are also described probabilistically. Prior to each development decision-making stage, the reservoir is operated using a closed-loop control approach, including model calibration against dynamic data, subject to geologic and future development and operation uncertainties. After each data assimilation step, a new stochastic optimization is performed to adjust controllable decision variables for the current well configuration (e.g., well rates or BHPs) using the updated models and potentially revised future development scenarios. Using a multistage stochastic optimization workflow, this process is repeated after each decision stage. Several numerical experiments are presented to discuss various aspects of the proposed closed-loop stochastic optimization formulation and to evaluate its robustness.