Babadagli

The 10 papers selected to appear in the August 2012 issue of SPE Res Eval & Eng focus on the areas of unconventional gas, EOR, and reservoir characterization.

Unconventional Gas

Gas Permeability of Shale presents a new approach to obtain the permeability of shale. The main observation is that laboratory measurements conducted with N₂ overestimate the permeability compared with CH₄. The main reason behind this is the CH₄ adsorption on the organic materials in nanoscale pores, resulting in permeability reduction. The paper investigates this phenomenon for different pressures and pore networks characteristics.

The next paper, Hydrate Decomposition and Its Material Balance in a Volumetric Tilted Hydrate-Capped Gas Reservoir by Method of Depressurization, reports an analytical model to predict the hydrate recovery of inclined hydrate reservoirs when produced at a constant rate. The model couples the solution of 1D heat flow from surrounding areas toward the decomposing zone, the tank-type material-balance equation, and the thermodynamic relation of hydrate decomposition. In addition to a series of sensitivity studies, an uncertainty analysis was performed for a hydrate reservoir in Alaska. The major observation was the undesirable acceleration of reservoir cooling at higher production rates, which is needed for faster recovery. This requires the optimization of operating conditions to avoid freezing.

EOR

Detailed Modeling of the Alkali/Surfactant/Polymer (ASP) Process by Coupling a Multipurpose Reservoir Simulator to the Chemistry Package PHREEQC  reports an improvement into numerical modeling of ASP flooding to account for the geochemistry involved and for the saponification process. This was achieved by coupling a multipurpose reservoir simulator with a geochemistry software program (PHREEQC) developed to calculate the equilibrium concentrations of chemical species. PHREEQC includes a rich database of chemical species and serves as a chemical reaction engine to determine the equilibrium state of the process modeled. The generic coupling process can be used for other chemical-based EOR methods.

The feasibility of air-foam flooding in waterflooded light-oil reservoirs was investigated experimentally in the paper Air-Foam-Injection Process: An Improved-Oil-Recovery Technique for Waterflooded Light-Oil Reservoirs. The effects of clay minerals and foam on low-temperature oxidation were clarified through combustion experiments. The main conclusion is that clay minerals could accelerate the low-temperature-oxidation reaction, while foam has a negative effect on this. By plugging gas/water channels from previous injection history and thereby improving the sweep efficiency, foam yielded an additional 10.9% recovery.

In the last paper of this category, Alkaline Steam Foam: Concepts and Experimental Results, the authors reported another experimental work on the use of foam at elevated temperatures. The tests used an alkali-surfactant system consisting of AOS 1618 and Na₂CO₃ for heavy Californian oil. The results revealed that the steam-foam process can be improved by injecting an alkali-surfactant mixture in the aqueous phase of the steam. The alkali injected reduced the surfactant adsorption and improved surfactant propagation. Significant reduction in residual oil saturation by the addition of alkali into the aqueous phase was reported. Also, tertiary recovery tests showed that the residual oil saturation can be as low as 0 to 5% with AOS 1618 and Na₂CO₃ in addition to steam, while the residual saturation with steam was limited to 13%.

Reservoir Characterization

The first paper under this category, A Modern Method for Using Databases To Obtain Accurate Solutions to Complex Reservoir-Characterization Problems proposes a new inversion method that predicts reservoir properties without using idealized model equations or minimization. After dividing the calibration database into input measurements and outputs (predicted reservoir properties), the new method maps the outputs using model-independent mapping functions constructed from Gaussian radial basis functions. Once the coefficients of the mapping function are determined using the database, there are no adjustable parameters. The new method is applicable to a wide variety of reservoir characterization problems if a database is available and it overcomes the inaccuracies of the conventional approaches.

Removal of Cyclic Borehole Noise From Low- and High-Resolution LWD Images and Its Impact on Image Interpretation presents an approach to improve the images obtained through logging-while-drilling (LWD) by removing borehole-oscillation noise using frequency-domain filtering. The application of this new algorithm to different field data illustrated considerable improvement in the images' quality. The processed images through this algorithm also provided a better description of bed boundaries in irregular boreholes and small-fracture detection. The method is not only applicable to density, but also to resistivity and ultrasonic images.

The next paper, Effect of Discontinuous Microfractures on Ultratight Matrix Permeability of a Dual-Porosity Medium introduced an analytical model that couples the composite matrix flow with the flow in a network of macrofractures, as in the conventional dual-porosity idealizations. The authors observed that the triple-porosity model is not sufficient to incorporate both micro- and macrofractures and vugs. It was also concluded that microfractures' contribution to the production could be more than expected in multiple fractured horizontal wells in ultratight formations, which entails critical effort in the characterization of microfractures.

Semianalytical modeling of a horizontal well in a homogeneous, dual, or triple porosity reservoir is the subject of New Type Curves for Modeling Productivity of Horizontal Well With Negative Skin Factors. By taking a negative skin approach and applying different solutions (e.g., Laplace transform, separation of variables, and inverse Laplace transform), the authors obtained a unified formula for both constant rate and constant pressure production, verified by real field data. A series of new standard log-log type curves were provided and different flow periods (e.g., early radial flow, hemi-radial flow, linear flow, and late pseudo-radial flow) were identified. Finally, the work was extended to horizontal well production in dual and triple porosity reservoirs, and V- and W-shaped derivative curves were obtained for dual and triple porosity systems, respectively.

In the paper A Comparison of Stochastic Data-Integration Algorithms for the Joint History Matching of Production and Time-Lapse-Seismic Data, the authors carried out a comprehensive comparison of three stochastic optimization methods in solving the 4D seismic history-matching inverse problem using synthetic and field data: particle swarm optimization (PSO), very fast simulated annealing (VFSA), and neighborhood algorithm (NA). Having applied these methods to the joint history-matching of 4D seismic and production data for an offshore West African reservoir undergoing waterflooding, they concluded that the PSO is more effective than the NA and VFSA methods for field applications. Although it provides a better framework for computing approximate posterior probability distributions of the history-matched model parameters, its computational cost is relatively higher than that of VFSA. The VFSA method is preferred when moderate computing resources are available.

Tayfun Babadagli,
University of Alberta