Streamlines Guided Assisted History Matching Methodology: A New Approach

Abstract

Abstract Accomplishing an acceptable history match of a reservoir simulation model with historical field performance data is a challenging task. The resources needed for assisted history matching for giant reservoirs is very demanding. Providing engineers with an optimized workflow is essential for these giant simulation models to accomplish the desired history match objectives in a timely manner. Reservoir engineers typically use one of many available commercial packages, some of them are stochastic in nature. These methodologies require users to define regions to guide the optimizer where to apply the modifiers. It is difficult to define a modifier region that can be used to relate the physics and fluids movements in the reservoir while preserving key geological signatures. All these concerns, if addressed properly can contribute to better understanding of the reservoir dynamics, reduce turnaround time of the project, improve quality of the history match, and optimize resources utilization. The work presented in this paper proposes a practical workflow to assist reservoir engineers identify important flow communications regions in the reservoir using streamlines properties. Many Producer-Injector pair regions are identified using streamlines, and an efficient strategy is used to select a handful of regions to apply permeability modifiers, for example. This workflow has been validated by history matching one of the giant fields in Saudi Arabia. A commercial history matching package and a customized streamlines tracing tool was used in this study. The results show improvements in the history match quality and reduction of overall history matching time with the current streamlines-guided workflow when compared to the conventional workflows that use the commercial package alone. The paper shows the impact of streamlines map choices on the quality of history match, turnaround time of history matching workflow, number of required simulation runs to complete a study and other relevant measures.