Truncated Multi-Gaussian Pore-Throat-Size Decomposition and a New Universal J-Function for Rock Characterization of Complex Carbonate Reservoirs

Abstract

AbstractRock-pore-space geometry and network topology have a great impact on dynamic reservoir characteristics, in particular on capillary pressure and relative permeability curves. In complex and heterogeneous carbonate reservoirs, the rock-pore-throat-size distribution is typically multimodal and its decomposition may be an important concept for reservoir characterization and simulation. Our objectives are to enhance current dynamic reservoir characterization processes and to assess the corresponding improved fidelity of reservoir simulation.We propose an innovative and precise mathematical formulation of the pore-throat-size distribution as a truncated multi-Gaussian decomposition, matching its parameters to experimental data through both direct and inverse approaches. Based on the above technique we also propose a new multi-Gaussian universal J** -function and a new dynamic reservoir-rock-typing index. We also prepare a comprehensive sensitivity study on the impact of multimodal dynamic reservoir characteristics on oil recovery results. For this sensitivity study, we use a synthetic dual-porosity carbonate-reservoir simulation model, incorporating the concept of the proposed universal J** -function to derive the corresponding dynamic reservoir properties.We applied the multi-Gaussian decomposition to the experimental mercury-injection-capillary-pressure (MICP) data from the Worldwide Rock Catalog (WWRC), provided by a joint-industry project (Core Lab, 2014). We used the multi-Gaussian universal J** -function to demonstrate the excellent results of such decomposition. These results showed large variability of multimodality for both clastic and carbonate rocks. We also found correlations between decomposition parameters, rock properties and rock textures. Considering the truncation of the pore-throat-size distribution proved to be important to model the physical and experimental limits of the data. The results of the sensitivity study showed a significant impact of multimodality heterogeneity on oil in place and reserves estimates for improved oil recovery (IOR) and enhanced oil recovery (EOR) processes in complex carbonate reservoirs, such as the ones found in the Brazilian Pre-Salt. It also showed the importance of considering pore-throat-size multimodality when deriving dynamic reservoir properties from capillary pressure and relative permeability experiments. In complex carbonate reservoirs under IOR/EOR, overlooking the rock-pore-space geometry and network topology may result in significant deviations in the quality of reservoir characterization and simulation results. In this context, proper multi-Gaussian decomposition and the introduction of a new multi-Gaussian universal J** -function are therefore crucial for carbonate reservoir simulations.The proposed truncated multi-Gaussian pore-throat-size decomposition presents significant additional benefits when compared to Thomeer's method. It also improves dynamic-reservoir-rock-typing and reservoir simulation processes. The new universal J** -function can be used to reconstruct capillary pressure curves from the information provided by multi-Gaussian pore-throat-size decomposition. Therefore, the new concepts presented in this paper have a clear potential to enhance the simulation of IOR and EOR in complex carbonate and clastic reservoirs.