PyFrac: A planar 3D hydraulic fracture simulator
Haseeb Zia, Brice Lecampion
Abstract
Fluid driven fractures propagate in the upper earth crust either naturally or in response to engineered fluid injections . The quantitative prediction of their evolution is critical in order to better understand their dynamics as well as to optimize their creation. We present an open-source Python implementation of a hydraulic fracture growth simulator based on the implicit level set algorithm originally developed by Peirce & Detournay (2008). This algorithm couples a finite discretization of the fracture with the use of the near tip asymptotic solutions of a steadily propagating semi-infinite hydraulic fracture. This allows to resolve the multi-scale processes governing hydraulic fracture propagation accurately, even on relatively coarse meshes. We present an overview of the mathematical formulation, the numerical scheme and the details of our implementation. A series of problems including a radial hydraulic fracture verification test, the propagation of a height contained hydraulic fracture, the lateral spreading of a magmatic dyke and an example of fracture closure are presented to demonstrate the capabilities, accuracy and robustness of the implemented algorithm. Program title: PyFrac CPC Library link to program files: http://dx.doi.org/10.17632/gv7yy9mmwj.1 Licensing provisions: GPLv3 Programming language: Python Nature of problem: Simulation of the propagation and closure of a planar three-dimensional hydraulic fracture driven by the injection of a Newtonian fluid in a material having heterogeneous fracture toughness under a non-uniform in-situ stress field. Solution method: The fully coupled hydro-mechanical moving boundary problem is solved combining a finite volume scheme for lubrication flow with a boundary element method for elasticity. The algorithm couples a finite scale discretization of the fracture with the near-tip asymptotic solution of a steadily moving hydraulic fracture. The fracture front is tracked via a level set approach using a fast marching method.