Abstract
The DFIT flowback analysis (DFIT-FBA) method, recently developed by the authors, is a new approach for obtaining minimum in-situ stress, reservoir pressure, and well productivity index estimates in a fraction of the time required by conventional DFITs. The goal of this study is to demonstrate the application of DFIT-FBA to hydraulic fracturing design and reservoir characterization by performing tests at multiple points along a horizontal well completed in an unconventional reservoir. Furthermore, new corrections are introduced to the DFIT-FBA method to account for perforation friction, tortuosity, and wellbore unloading during the flowback stage of the test.
The time and cost efficiency associated with the DFIT-FBA method provides an opportunity to conduct multiple field tests without delaying the completion program. Several trials of the new method were performed for this study. These trials demonstrate application of the DFIT-FBA for testing multiple points along the lateral of a horizontal well (toe stage and additional clusters). The operational procedure for each DFIT-FBA test consists of two steps: 1) injection to initiate and propagate a mini hydraulic fracture and 2) flowback of the injected fluid on surface using a variable choke setting on the wellhead. Rate transient analysis methods are then applied to the flowback data to identify flow regimes and estimate closure and reservoir pressure. Flowing material balance analysis is used to estimate the well productivity index for studied reservoir intervals.
Minimum in-situ stress, pore pressure and well productivity index estimates were successfully obtained for all the field trials and validated by comparison against a conventional DFIT. The new corrections for friction and wellbore unloading improved the accuracy of the closure and reservoir pressures by 4%. Furthermore, the results of flowing material balance analysis show that wellbore unloading might cause significant over-estimation of the well productivity index. Considerable variation in well productivity index was observed from the toe stage to the heel stage (along the lateral) for the studied well. This variation has significant implications for hydraulic fracture design optimization, particularly treatment pressures and volumes.
A Hybrid Artificial Intelligence Model to Predict the Elastic Behavior of Sandstone Rocks