Determination of Effective Parameters of Gas Injection in Naturally Fractured Reservoirs by Combination of Reservoir Simulation and Design of Experiment Techniques

Abstract

Screening analysis is a useful guideline which helps us with proper field selection for different enhanced oil recovery processes. In this work, reservoir simulation is combined with experimental design to estimate main effects and possible interactions of reservoir rock and fluid properties on performance of different gas injection processes in naturally fractured reservoirs (NFRs). Studied parameters include reservoir thickness (h), oil viscosity μo), pore size distribution (λ), horizontal permeability (Kh), storage capacity (ω), reservoir dip, critical water saturation (Swc) and threshold capillary pressure (pct). The recovery factors of different simulation designs are analyzed by use of fractional factorial design (FFD) approach. Finally, the statistical significance of results are evaluated by hypothesis testing and ANOVA, and presented by Pareto and tornado plots. Main effect analysis showed that effective parameters are ordered with respect to their importance as follows: For Methane and nitrogen injection: Kh, dip, Swc, ω, h, μo and λ; threshold capillary pressure showed a minor effect on recovery factor. For Carbon dioxide injection: μo, dip, Swc, ω, h and Kh; Pore size distribution (λ) and threshold pressure were not shown to be statistically significant in this process. Likeness of main effects and parameter interactions for methane and nitrogen injection prove the similarity of dominant mechanisms and characteristics of these processes compared to CO2 injection which seems almost different.