Reservoir Simulation Model for Fractured and Partially Fractured Reservoirs based on PEBI Grids

Abstract

This paper presents the implementation of the dual continuum approach (dual porosity/permeability), into a PEBI grid based, general-purpose reservoir simulator. Experience shows that not all parts of a reservoir will typically require dual porosity/permeability treatment. Sometimes numerical or other restrictions limit the use of a dual porosity/permeability model. The approach presented in this paper takes these aspects into consideration. It allows creating dual porosity/permeability cells wherever the geological attributes (facies, rock type, fracture intensity) imply to do so. The remaining - unfractured - part of the model will be treated like conventional single porosity areas. Therefore, the number of grid blocks and the CPU time required for the flow simulation are minimized. Fracture and matrix-fracture transfer properties (including shape-factors) are derived from the geological information and fracture intensity. All of this information and data is stored on a cell by cell basis. Furthermore, the presented method does not restrict any of the superior gridding capabilities of the simulator like local grid refinement, fault or well modeling. The equation system resulting from all grid blocks will be solved fully coupled. The flow simulator is based on a multi-purpose, compositional formulation, and copes with systems ranging from single-phase to multi-phase, multi-component fluid flow. Several examples illustrate the usefulness and flexibility of this modeling approach for accurate and economic simulation of fractured or partially fractured reservoirs. Introduction The existence of fractured reservoirs is known since many years. In fact, one may claim that all hydrocarbon reservoirs are fractured. The question remains, however, whether or not those fractures are important for fluid flow (i.e. form a fracture network) and if their existence justifies use of a dual porosity approach in the simulation. Only when fractures with sufficient length of penetration, connectivity and spacing occur, do their effects become important. To assess fracture contributions accurately, a wide range of geological, petrophysical and engineering determinations are required on large as well as small scales and their complex interaction must be investigated. A classification of fractures is typically based on either descriptive criteria (natural fractures vs. induced fractures, macro fractures and micro fractures, open and closed fracture) or geological criteria (origin of fractures, stress state, folding, stratigraphy, etc.). For the purpose of reservoir simulation, the classification after Nelson1,2 or modifications of it3 are used most commonly. This classification defines the following types of fractured reservoirs:Type 1: Fractures provide the essential porosity and permeabilityType 2: Fractures provide the essential permeabilityType 3: Fractures provide permeability assistance to an already producible reservoir Only with the first type is the accurate calculation of fracture porosity, fracture width and spacing of paramount importance. The estimation of these properties will predict whether initial flow rates can be maintained or will drop rapidly within short time. With this type of fractured reservoirs, the producible hydrocarbon fluid volumes reside within the fracture network and the application of a dual porosity approach must be questioned. For reservoir types 2 and 3, where the main storage volume resides in the matrix rock volume, early knowledge of matrix-fracture interaction is extremely important. It will determine how efficiently the matrix can be drained by the fracture system. The accurate determination of the fracture porosity will be of less significance in those cases.

Simulation of Naturally Fractured Reservoirs. State of the Art

“…Recent developments in fractured reservoir simulation concern the possible impact of geomechanical parameters on the fracture flow properties, porosity and permeability (Chen and Teufel, 2000;Bagheri and Settari, 2008), the need of high-performance simulators with the use of parallelization techniques for giant fractured field simulation (Al-Shaalan et al, 2003) and the use of streamlines for waterflood simulations with three forms of the transfer function (Di Donato et al, 2003), and also the implementation of PEBI grids in order to keep a better track of the actual/geological fault/fracture network than conventional dual-porosity models (von Pattay and Ganzer, 2001). …”

Recovery performance of partially fractured reservoirs by capillary imbibition

“…In this case, the fracture network does not spread over the entire reservoir volume, i.e., only parts of the reservoir are fractured. Such reservoirs are known as Partially Naturally Fractured Pattay and Ganzer (2001) presented a simulation model for PNFR based upon perpendicular bisector grids (PEBI). They proposed that with the help of PEBI grid technology, different areas in the reservoir simulation model can be treated selectively and assigned fractures, matrix, and matrix-fracture exchange parameters on a cell-by-cell basis.…”