ABSTRACT
Modeling frameworks for fractured geosystems aims for high accuracy in representing the subsurface fluid flow mechanisms in complex natural fractured reservoirs. However, modeling a reliable fractured reservoir model can be very challenging due to myriads of uncertainties in the scaling-up process. This study compares Dual Porosity and Dual Permeability (DPDK) against Embedded Discrete Fracture Model (EDFM) method in terms of fluid volume recoveries and simulation time. DPDK model was constructed by using single phase flow, and upscaling discrete fractures geometries. On the other hand, the same scenario was recreated by EDFM method to translate the mentioned fractures into numerical reservoir modeling without the need of upscaling the grid properties. Then, the dynamics of the reservoir were simulated, and the impact of the fractures in each framework were quantified. After analyzing the matrix-fracture and fracture-fracture flows, the findings include that EDFM technology strictly models the fracture geometries and does not undertake interconnection of fractures that are very close to each other unless they are explicitly connected, which differs from DPDK framework because the latter one does assemble these very-close fractures after upscaling.
Fractures and faults in gas and petroleum reservoirs have had a critical impact on the dynamics for the assessment of field productivity and reserves. Modeling these fracture systems represents a great challenge if upscaling their properties, geometries, orientations and intensities cannot be accurate. Likewise, the clustering of effective fractures is often difficult to emulate, triggering fractures to be often randomized or arbitrary leading to not-so-realistic characterization (Tavakkoli, 2009; Koike et al., 2012; Wang and Vecchiarelli, 2019). Consequently, it is necessary to evaluate the fidelity of natural fracture modeling frameworks to seize their impact on the behavior of these fractured reservoirs for proper management and economic development.
Widespread use of numerical frameworks to model and upscale naturally fractured reservoirs (NFR) include continuum approaches such as Dual porosity (DP) approach or Dual Porosity and Dual Permeability (DPDK) method. The equivalent continuum approach converts complex and irregular fracture geometry systems into regular spaced 3D fracture sets in the macroscopic level. For instance, DPDK intends to characterize explicitly the petrophysical properties of the DFN in order to obtain equivalent parameters for dynamic fluid flow models by employing regularly spaced matrix blocks separated by a fracture network (Warren and Root, 1963; Kazemi et al, 1976; Thomas et al., 1983; Uleberg and Kleppe, 1996) (See Figure 1). However, this method cannot accurately deal with complex fracture geometries and tends to overestimate the transmissibilities. (Su et al., 2013; Vo et al., 2019). Therefore, modeling with DPDK method may not capture important fractured rock mechanisms accurately, provoking misleading flow estimates.
