So, what is a DFIT? A DFIT is a diagnostic fracture-injection/falloff test, and is considered the unconventional well test. A DFIT is also likely to be the only well test in the life cycle of an unconventional well. DFIT are designed to determine the following.
A DFIT consists of an injection of a relatively small volume of water (or gas) at a rate sufficient to create and propagate a hydraulic fracture. Injection volumes considered “small” depend on the available shut-in time and the desired sample size of the reservoir rock. For example, many DFIT with injection volumes exceeding 1,000-bbl have been pumped, but the shut-in period was in terms of weeks instead of hours or days. The analysis methodology is based on an "impulse" assumption that simply requires that the time of the injection be short relative to the reservoir response.
Provided a fracture is created during the injection, the injection volume controls the size of the reservoir rock sampled during a DFIT (Craig and Jackson, 2016). In very rare circumstances, a hydraulic fracture will not be created during a DFIT injection, but the falloff data can still be analyzed as a DFIT using the short-flow transient analysis workflow (Craig, 2014). A DFIT with an injection volume as small as one liter can be pumped, but the test may not sample sufficient reservoir rock to provide a representative effective permeability estimate (Craig and Jackson, 2016). There is always a trade-off between an adequate injection volume to obtain a representative transmissibility estimate and the available shut-in time.
Following the injection, the well is shut-in for an extended falloff period. Traditional DFIT analysis dating back to the late 1990s allowed the before-fracture-closure data to be analyzed for closure stress, leakoff type, pressure, and permeability. Before-closure estimates of pressure and permeability have the highest uncertainty of all the available analysis methods. The after-fracture-closure data could also be analyzed for reservoir pressure and transmissibility provided very specific flow regimes were observed during the falloff—bilinear, linear, radial. If radial, or pseudoradial, flow was observed, then an estimate of permeability could be made independent of the created fracture half length. All other flow regimes required an estimate of the fracture half-length open to flow during the falloff to estimate permeability. Unfortunately, fracture half-length open to flow is most often unknown or highly uncertain.
The well-test analysis workflow (Craig, 2014) developed by DFITpro allows all the falloff data from before-closure to after-closure to be analyzed using analytical pressure transient solutions. The well-test analysis workflow allows data to be analyzed for any reservoir and wellbore model — anisotropic, inhomogeneous, naturally fractured, a finite-conductive hydraulic fracture, multiple finite-conductivity hydraulic fractures, damaged hydraulic fracture(s), etc. Provided a solution can be programmed, the DFIT falloff data can be matched to a corresponding solution. Of course, typical well-testing nonuniqueness issues exist, but in most cases, the well-test analysis workflow will provide a reasonable interpretation when traditional DFIT methods cannot.
DFIT are not limited to prefrac testing. DFIT can also be used after significant production to define average reservoir pressure, hydraulic fracture half-length, conductivity, permeability, and fracture damage. In most post-production tests, nitrogen is injected instead of water, which allows surface gauges to be used to record the falloff and allows easier cleanup after the test. Nitrogen injections as part of a diagnostic fracture-injection/falloff test are referred to as NFIT (Tomkins, et al).
RNIFT or RIFT refer to refracture nitrogen injection/falloff and refracture injection/falloff tests, respectively. RNIFT or RIFT can be used to diagnose well conditions prior to implementing a refracturing program (Craig and Odegard, 2008). RNIFT and RIFT are a core technique used in DFITpro’s restimulation program and in a process we call restimulation-candidate IDs—Identify, Diagnose, and stimulate.