TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractThis paper describes the feasibility study of a large-scale miscible CO 2 -WAG (MWAG) injection scheme in the Gullfaks Field, offshore Norway. We describe the reservoir engineering workflow and simulation techniques, the predicted production and injection profiles, and the main infrastructure solutions under consideration.Compositional cross-section models and recently available streamline-tracer simulation techniques are employed to scale up from element models to a fast, full-field simulator with a high degree of flexibility. Figure 1: Gullfaks Field LocationThe starting point for the workflow is a set of black oil and streamline front tracking models, history matched on coarse and fine grids. A fast, finely gridded streamline model is used to identify the MWAG injection targets, define injection well locations and completion strategy. Fine gridded cross-sections are extracted and used in a compositional simulator to study and quantify the miscible displacement process. These are the used to derive scaling parameters used in a simple, ultra-fast streamline-tracer model, scaling the MWAG process up to field level. The streamline-tracer model interactively optimises solvent allocation and generates production predictions on a well-by-well basis. Water flood recovery and incremental IOR are predicted simultaneously in a single simulation run.In addition, the general economic limitations and example technical solutions for implementation of a CO 2 MWAG on the Gullfaks Field are briefly described.

Production from the structurally complex Gullfaks Field in the northern North Sea is now on decline, reduced by a third from the peak year 1994, when oil production exceeded 30 MSm3. Recoverable reserves are currently estimated at 319 MSm3, of which approximately 240 MSm3 have been produced to date. Approximately 80% of the recoverable reserves lie in the middle Jurassic Brent Group. An important goal for the Gullfaks license is to increase the recoverable reserves by 47 MSm3, of which the Brent Group isexpected to yield 30 MSm3 through a recently launched technology programme. Locating the remaining oil is an important part of this programme. Time lapse seismic data are considered to be a key element in this process, and a multidisciplinary team is now in the process of integrating 4D seismic data with other available reservoir and production information, in order to improve the static, as well as the dynamic, reservoir description. The project is also a testing ground for several recent R&D results in the field of geological modelling, reservoir and production engineering, assisted by modern visualization techniques to enhance communication of technical information and ideas. An example of how the multidisciplinary approach has improved our reservoir and drainage understanding will be demonstrated. This improved understanding provides a better basis for quantifying the field's IOR potential. Introduction Reservoir Description. The Gullfaks Field is located in the Norwegian sector of the North Sea, in block 34/10, approximately 175 km northwest of Bergen, where the Gullfaks Business Unit is based, on the west coast of Norway, see Figure 1. The reservoir units are sandstones of early and middle Jurassic age, around 2000m sub sea and measure several hundred meters thick (Ref. 1). The uppermost Brent sequence contains roughly 80% of the reserves, with the deeper Cook and Statfjord formations contributing the remainder. Reservoir quality is generally very high, with permeability ranging from few tens of mD to several Darcys depending on layer and location. Figure 2 shows a cross section indicating the quality and variability of the reservoirs. The reservoirs are over pressured, with an initial pressure of 310 bar at datum depth of 1850 m below mean sea level, and a temperature of 70 degrees C. The oil is undersaturated, with a saturation pressure of approximately 245 bar, depending on formation depth and location. The GOR ranges between 90 and 180 Sm3/Sm3, with stock tank oil gravity around 860 kg/m3. Structurally, the field can be divided into three regions(Ref. 2). The so-called 'Domino Area' with rotated fault blocks in the west, and a Horst area in the east. In between is a complex 'Adaptation Zone', characterized by folding structures. The north-south faults that divide up the field have throw up to 300 meters. In the western part the faults slope typically around 28 degrees downward to the east, whereas in the eastern horst they slope 60–65 degrees downwards to the west. The field is further cut by smaller faults, with throws of zero to few tens of meters, both in the dominant north-south as well as east-west direction. Many of these lesser faults have slopes of 50–80 degrees.

Every simulation engineer wishes to simulate large full field models, but historically reservoir simulation of the SAGD process has been constrained to single well models up to a single pad. Models of these sizes provide valuable information and have helped to assess the development potential of reservoirs. These models may be used for reservoir management and to support the decision making process for the design of the initial completion, operating strategy, multi-pad wind down evaluations and also qualitatively assess the uncertainty in the SAGD forecast under different geological settings.However, in many cases we are left with the question of how multi-well and multi-pad communications ultimately affect performance at the well pair scale. Due to technological constraints with computer hardware and simulation technology, running extremely large multi-pad models has been until recently largely impractical, especially when trying to run multiple scenarios to better understand the impact of geological and operational uncertainty.In this paper, we present a new and practical workflow that makes running extremely large multi-pad, multi-million grid cell SAGD models a reality. The three major steps of the workflow are 1) Generating simulation friendly geomodels, 2) Use of experimental design and 3D sub-models based on SAGD Performance Index (SPI) for numerical tuning, and 3) Use of 2D cross-sections and SPI to develop dynamic grid refinement parameter values for the full 3D model. All of these steps are aimed at improving the numerical stability and run time of multi-pad SAGD simulation models.A 24 SAGD well pair model with 2.52 million gridblocks was simulated for 10 years of forecast. The reservoir is geologically complex and highly heterogeneous. We discuss some of the important aspects that need to be accounted for when simulating large scale SAGD models. Using this new workflow, the simulation run time was reduced from 42 days to 7 days on 8 CPUs -a 6 times speedup. The resulting run time is short enough to facilitate multi-realization simultaneous runs using 8 CPUs hence maximizing the throughput and minimizing the simulation cycle time. This new workflow can be easily replicated and, more importantly, automated to reduce engineering time requirements.

No abstract

Summary Every simulation engineer wishes to simulate large full-field models, but historically reservoir simulation of the steam-assisted-gravity- drainage (SAGD) process has been constrained to single-well models up to a single pad. Models of these sizes provide valuable information and have helped to assess the development potential of reservoirs. These models may be used for reservoir management and to support the decision-making process for the design of the initial completion, operating strategy, and multipad wind-down evaluations, and also qualitatively assess the uncertainty in the SAGD forecast under different geological settings. However, in many cases we are left with the question of how multiwell and multipad communications ultimately affect performance at the well-pair scale. Because of technological constraints with computer hardware and simulation technology, running extremely large multipad models has been until recently largely impractical, especially when trying to run multiple scenarios to better understand the impact of geological and operational uncertainty. In this paper, we present a new and practical workflow that makes running extremely large multipad, multimillion-grid-cell SAGD models a reality. The three major steps of the workflow are (1) generating simulation-friendly geomodels, (2) use of experimental design and 3D submodels on the basis of SAGD performance index (SPI) for numerical tuning, and (3) use of 2D cross sections and SPI to develop dynamic grid-refinement-parameter values for the full 3D model. All of these steps are intended to improve the numerical stability and run time of multipad SAGD simulation models. A 24-SAGD-well-pair model with 2.52 million gridblocks was simulated for 10 years of forecast. The reservoir is geologically complex and highly heterogeneous. We discuss some of the important aspects that need to be accounted for when simulating large-scale SAGD models. Using this new workflow, the simulation run time was reduced from 42 days to 7 days on eight central processing units (CPUs)—a six-time speedup. The resulting run time is short enough to facilitate multirealization simultaneous runs using eight CPUs, hence maximizing the throughput and minimizing the simulation cycle time. This new workflow can be easily replicated and, more importantly, automated to reduce engineering time requirements. While this paper focuses on the SAGD process, this methodology is completely generic in that it can be applied to any large data set for any process. Details will differ depending on the process, but the workflow will be the same.