Brownfield Development Optimization under Uncertainty – A Structured Workflow Design for Complex Case Scenarios

Abstract

AbstractAn increasing number of field development projects include rigorous uncertainty quantification workflows based on parameterized subsurface uncertainties. Reservoir model calibration workflows for simulation studies including historical production data, also called history matching, deliver non-unique solutions and remain technically challenging. This work presents an integrated workflow design for brownfield development optimization projects under uncertainty. Challenges related to complex simulation models with long run times are addressed. Proxy modeling techniques are introduced for performance improvement with application to history matching and optimizing field developing scenarios. Selection strategies of multiple history-matched prediction candidates for estimating prediction uncertainties are presented.Workflow designs for history matching require scalable and efficient optimization techniques to address project needs. A structured workflow design is introduced for addressing complex project requirements and to define intermediate milestones for project reviews. Experimental design techniques are introduced for sensitivity analysis and dimension reduction. History matching processes are built on a series of parameter screening and derivative-free optimization techniques. We introduce Markov Chain Monte Carlo (MCMC) techniques for optimization and uncertainty quantification. Rejection filtering is used for selecting alternative prediction candidates in a multi-objective solution space.Field development optimization workflows define challenges due to the nature of a high-dimensional discrete option space. We present a combined usage of experimental design techniques for generating training data sets and proxy modeling techniques for extensively screening a multi-dimensional control parameter space. Visualization techniques are used to present a solution-frontier of optimized field development scenarios meeting multi-objective optimization criteria, e.g., economic field performance criteria as well as field production targets.The structured workflow was designed and applied to an undisclosed field development optimization project. In this paper we describe the motivation and benefit of experimental design and optimization techniques with application in various phases of the structured workflow from history matching to field development optimization under uncertainty. Complexities in prioritizing selection criteria of cross-disciplinary static and dynamic uncertainty parameters included in a multi-dimension uncertainty matrix are discussed under practical integration criteria. Performance criteria for workflow deliveries in time and quality are critical in decision processes for moving forward within the project phases and for delivering supportive results for reservoir project management. We present decision criteria at which workflow stages added computation resources are beneficial and which methods are scalable and support distributed computing-technology for reducing elapsed project time.