This paper presents a comprehensive approach to the design of hydraulic-fracture treatments, accounting for anisotropic stress conditions, rock properties, and the effect of pore-pressure changes caused by production in tight gas reservoirs. This has allowed us, among other opportunities, to design a refracture treatment. The poroelastic model is also coupled with a production-optimization scheme to optimize the design parameters for hydraulic-fracture treatments. A case study of refracture treatment has been carried out for a typical tight gas reservoir. This study has shown that the fracture treatment can be optimized successfully to increase the net present value and/or ultimate gas recovery. This study also has demonstrated that a second fracture treatment can be performed after a period of production from the same treated interval to maintain production without the drilling of additional wells.
Traditionally, designing a fracture treatment entails a 3-step procedure: (1) determination of the fracture geometry on the basis of a given set of treatment parameters, (2) estimation of production from the designed fracture geometry, and (3) estimation of net present value for the designed treatment. A set of treatment parameters that gives the highest net present value is considered to be the optimum treatment design. This procedure, however, does not account for events that occur over the production life of the treated well: low reservoir pressure, proppant degradation, or embedment that results in severe fracture-conductivity impairment. Our approach seeks to provide a remedy for these problems by optimizing the fracture treatment and maximizing net present value for a given reservoir condition.
The approach makes use of production-induced reservoir stress changes. This phenomenon has been observed both in the field (Wright and Conant 1995) and in the laboratory (Bruno and Nakagawa 1991). Previous studies suggest that with pore-pressure depletion, the effective stress orthogonal to the fracture changes faster than one along the fracture, causing stress reversal. This stress reversal could be exploited to improve reservoir productivity by means of a refracture treatment (secondary fracture treatment). The secondary fracture created at this stage propagates in a direction different from that of the initial fracture. This refracture treatment has the potential to increase production by intersecting undrained areas.
In recent years, application of oriented refracturing has been gaining attention. Production tests and history matching, as well as downhole and surface tiltmeter measurements, show that a secondary fracture, under certain conditions, can reorient up to 90° relative to an initial hydraulic fracture. A schematic of fracture reorientation is presented in Fig. 1.
Production from the initial fracture causes a local depletion of pore pressure around the wellbore and the fracture. Because of poroelastic effects, the pore-pressure depletion changes stresses in the reservoir (Biot 1941, 1956; Geertsma 1957; Raghavan and Miller 1975; Rice and Cleray 1976; Verruijt 1969). The horizontal stress component parallel to the initial fracture (maximum horizontal stress) is reduced more rapidly than the perpendicular component (minimum horizontal stress). If the induced change in stress overcomes the initial stress differential, then the direction of the minimum horizontal stress becomes the direction of the maximum horizontal stress (stress reversal) around the wellbore. Studies have shown that stress reversal is more pronounced in regions with high anisotropic horizontal permeability (Siebrits et al. 1998). Under these conditions, a secondary fracture can be initiated and propagated along a different azimuth plane (up to 90° from the initial fracture) (Elbel and Mack 1993; Siebrits et al. 1998). The fracture may continue to propagate along the new azimuth for some distance beyond the isotropic boundary (see Fig. 1), depending on formation toughness. Note that the stress changes reach their maximum value and then diminish with further pore-pressure depletion (Siebrits et al. 1998). Thus, an optimal time window can be obtained to carry out a potential secondary fracture treatment.
A holistic approach that takes advantage of stress changes induced by production operations in the design of secondary fracture treatments is used. Four models were used for this purpose: (1) poroelastic reservoir model, (2) fracture-geometry model, (3) production model, and (4) economic model. In the following section, we describe the poroelastic model and conduct a sensitivity analysis to show how different parameters affect refracture treatments. Following this, an optimization technique to design an optimum refracture treatment for a tight gas reservoir is presented. The optimization technique combines the fracture-geometry model, the production model, and the economic model.