Streamline-Based Production Data Integration in Naturally Fractured Reservoirs

Abstract

Abstract Streamline-based models have shown great potential in reconciling high resolution geologic models to production data. In this paper we extend the streamline-based production data integration technique to naturally fractured reservoirs. Describing fluid transport in fractured reservoirs poses additional challenge arising from the matrix-fracture interactions. We use a dual-porosity streamline model for fracture flow simulation by treating the fracture and matrix as separate continua that are connected through a transfer function. Next, we analytically compute the sensitivities that define the relationship between the reservoir properties and the production response in fractured reservoirs. The sensitivities are an integral part of our approach and can be evaluated very efficiently as 1-D integrals along streamlines. Finally, the production data integration is carried out via a generalized travel time inversion which has been shown to be robust because of its quasi-linear properties and utilizes established techniques from geophysical inverse theory. We also apply the streamline-derived sensitivities in conjunction with a dual porosity finite difference simulator to combine the efficiency of the streamline approach with the versatility of the finite difference approach. This significantly broadens the applicability of the streamline-based approach in terms of incorporating compressibility effects and complex physics. We demonstrate the power and utility of our approach using 2-D and 3-D synthetic examples designed after actual field conditions. The reference fracture patterns are generated using a discrete fracture network (DFN) model that allows us to include statistical properties of fracture swarms, fracture densities and network geometries. The DFN is then converted to a continuum model with equivalent grid block permeabilities. Starting with prior models with varying degrees of fracture information, we match the water-cut history from the reference model. Both dual porosity streamline and finite difference simulators are used to model fluid flow in the fractured media. Our results indicate the effectiveness of our approach and the role of prior information and production data in reproducing fracture connectivities and preferential flow paths. Introduction Natural fractures are known to play a significant role in subsurface flow and transport of fluids. In recent years, advances in key technologies such as seismic imaging and horizontal drilling revealed the true extent of fractures in many reservoirs and enabled operators to utilize novel ways to use fracture connectivity to enhance recovery. The number of reservoirs that are now considered to be naturally fractured has also risen significantly in recent years and there is a greater need for more robust fracture characterization methods that can integrate both static and dynamic data in an efficient manner.1 Of late, discrete fracture network (DFN) techniques have gained increasing attention in the oil industry.2,3 The DFN is based on mapping fracture planes in 3D space using statistical properties of fracture swarms, fracture network geometry and flow characteristics. The advantage of the DFN models is the ability to incorporate complex fracture patterns based on field data such as cores, well logs, borehole images, seismic data and geomechanics. Although the DFN models can reproduce very realistic fracture geometry, it is important to condition these models to dynamic data such as well test, tracer and production data to reproduce the flow behavior in the reservoir. Such conditioning is particularly important for fractured reservoirs because only a small fraction of the fractures in the DFN model might carry bulk of the fluid flow.4,5 Streamline models have shown great potential in integrating dynamic data into high resolution geologic models.6–10 A unique feature of streamline models has been the ability to efficiently compute the sensitivity of the production data to reservoir parameters such as porosity and permeability. These sensitivities are partial derivatives that quantify how the production response will be affected by changes in reservoir properties. Integrating dynamic data into reservoir models typically involve the solution of an inverse problem and the sensitivities play a key role here. In our previous works, we have utilized the streamline-based sensitivities in conjunction with a generalized travel time inversion method to efficiently integrate production data into geologic models.7 Our approach has been successfully applied to a large number of field cases including a giant middle-eastern carbonate reservoir.8