51st U.S. Rock Mechanics/Geomechanics Symposium,
San Francisco, California, USA
2017. American Rock Mechanics Association
2 in the last 30 days
135 since 2007
Show more detail
ARMA Member Price:
ARMA Non-Member Price:
ABSTRACT: We explore the feasibility of using a bonded-particle model (BPM) to simulate the rock disaggregation process occurring near a wellbore perforation in sandstone by addressing the simpler question: Can we construct a 2D BPM that produces surface fragments when subjected to boundary conditions similar to those around a wellbore perforation in dry sandstone? The answer to this question is yes. We can construct a 2D flat-jointed material to represent Castlegate sandstone. Our synthetic material matches much of the macroscopic response (including the direct-tension and unconfined-compressive strengths) and many of the mechanisms that occur during direct-tension and compression tests as well as the trends in the macroscopic response and the primary mechanism that occurs during thick-walled cylinder (TWC) tests to produce a breakout failure type. The primary mechanism is termed “buckling-assisted fragmentation,” in which a buckling and spalling process produces thin fragments of rock similar to onion skins, and thereby produces a breakout failure type. The perforation-collapse behavior of our synthetic material is related to the hole resolution (defined as the number of grains across the borehole diameter), with TWC strength decreasing as hole resolution increases. This observation suggests that perforation strength in a given material will decrease with increasing perforation size.
The sand-production process that occurs near a wellbore perforation in competent sandstone in response to hydrocarbon production can be described by a two-stage conceptual model in which the perforation becomes filled with detached fragments during the first stage, and with further drawdown during the second stage, these fragments become smaller. The first stage of the process is characterized by properties obtained from laboratory tests (such as tensile, triaxial and thick-walled cylinder) and assessed by calibrated continuum-based sand-prediction models, but the second stage of the process is difficult to characterize and assess quantitatively. The second stage has been assessed via empirical equations linking the rock strength degradation to production conditions (Xiao and Vaziri, 2011). A long-term goal of the present work is to quantify the entire rock strength degradation process by employing a bonded-particle model (BPM, Potyondy and Cundall (2004); Potyondy ) in which the relevant grain-scale processes are modeled directly so that the rock strength degradation becomes an emergent property of the model arising from the microstructural interactions of the grains and cement.
Number of Pages
Looking for more?
Some of the OnePetro partner societies have developed subject- specific wikis that may help.