Abstract

Abstract Fractured reservoirs, be it naturally fractured ones or hydraulically fractured, are an important asset class of hydrocarbons. In naturally fractured reservoirs (NFRs) and unconventional reservoirs (UCRs) with hydraulic fractures, the rock matrix and fractures exhibit a large contrast in their conductivity to fluid transport limiting the ability to enhance recovery by traditional viscous forces mechanisms. During IOR processes such as gas or water injection, NFRs may also contain different fluid phases with different densities making the gravity and capillary forces relatively significant to impact the incremental production. The oil recovery mechanisms from Naturally Fractured Reservoirs (NFRs) and Unconventional Reservoirs (UCRs) - oil expansion, imbibition and gravity drainage, can be modeled using the Dual Porosity Dual Permeability (DPDK) or Embedded Discrete Fracture Model (EDFM) options in reservoir simulation. The main differences in the dual porosity and discrete fracture models are in the early time flow behavior as the saturation changes is modeled with different levels of efficiency and accuracy. NFRs represented as Dual media models (either dual or single permeability) use pseudo functions to represent gravity drainage. These models are limited by the assumptions used while deriving these coupling models between the different media. The higher-level resolution of Discrete Fracture-Matrix (DFM) has been previously reported to better represent the physics of these processes (water or gas injection). EDFM models simulate gravity drainage explicitly. Capillary forces driving imbibition in matrix is modeled using capillary pressure curve in both EDFM and DPDK models. In this paper we benchmark the behavior of both the DPDK and EDFM approaches and their results are compared to provide recommendations for fractured reservoir simulation modeling. For the unconventional reservoirs (UCRs), due to very large permeability and porosity variation between reservoir rock and hydraulic fractures, dual porosity representation is not suitable to capture phase transitions and near well-bore effects. The paper highlights the limitations of DPDK modeling methods and comparison is presented with DFM methods for such reservoirs.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call