Interpretation of pseudotransient linear flow in production data analysis of tight, naturally fractured reservoirs

Abstract

This dissertation delineates a complete set of flow regimes for fractured horizontal wells in unconventional reservoirs. Nine flow regimes are identified from asymptotic approximations of the trilinear flow model for fractured horizontal wells surrounded by a stimulated reservoir volume (SRV). The transient dual-porosity slab idealization is used to define fluid transfer from matrix blocks to natural fractures in the SRV and the production of an isothermal, single-phase, and slightly compressible fluid is considered at a constant bottomhole pressure. The solution is in the Laplace transform domain and numerically inverted to real-time domain to obtain the production rates as a function of time. Although the model and the results presented in this work can be extended and applied to fractured horizontal wells in shale-gas plays, the assumption of transient dual-porosity slab reservoir may not a be good approximation due to the complex fracture networks in these systems. On the other hand, the model is a good proxy for tight-oil plays, such as the Eagle Ford, Bakken, and Niobrara, where the thin producing layers of carbonates are interbedded by layers of tight shale, which acts as the storage medium. The main contribution of this work is the identification of a new pseudotransient linear flow regime, Flow Regime 6, and the description of the appropriate production data analysis procedures. Pseudotransient linear flow has the conventional diagnostic features of linear flow (½-slope straight line on log-log coordinates) but corresponds to the physical conditions (i) where the flow in the matrix is pseudosteady while the natural fractures have transient flow, which is referred to as Flow Regime 6a, or (ii) where the natural fractures experience boundary dominated flow concurrent with transient flow in the matrix, called Flow Regime 6b in this work. Flow Regime 6a is analogous to the infinite-acting total system (matrix and fracture) flow in conventional transient, dual-porosity slab models and the analysis yields the product of natural fracture flow capacity and storativity if the fracture half-length is known or vice versa. Flow Regime 6b, on the other hand, is possible in ultra-tight fractured unconventional reservoirs and has not been reported in the literature. If it is identified, analysis yields the volume of the natural fractures provided that the permeability and storativity of the matrix is known or vice versa. Because transient and pseudotransient linear flows exist at early and late times, respectively, it may be possible to make a reasonable assertion whether the linear flow behavior observed on the data corresponds to transient or pseudotransient linear flow period. However, both Flow Regimes 6a and 6b exist in the same late-time periods and cannot be labeled a priori. This has serious repercussions on the interpretations and properties that can be derived from the analysis. The diagnostic features and time ranges of existence given in this dissertation are important tools to identify and distinguish between Flow Regimes 6a and 6b.