Abstract
This paper covers the methodology to derive all geomechanical properties (Young's modulus, Poisson's ratio and vertical/horizontal variable Biot constants as a function of rock type) for 13 different stress models. Minimum horizontal stress (Sh) is a key parameter controlling fracture height growth during hydraulic fracturing simulation. Assuming a homogeneous formation (rock property Horizontal:Vertical = 1.0) or poorly derived inputs for the anisotropy model can lead to incorrect fracture geometry. A major assumption made using the various stress models is the Biot poro-elastic constant. Many default models assume a Biot poro-elastic constant of one, which is valid for coarse grained conventional reservoirs where porosity is greater than 20%. Most of the reservoirs stimulated with hydraulic fracturing today do not fall in that porosity range, therefore an alternative derivation for the Biot poro-elasticity and its variability requires additional discussion.
Models derived and compared with their associated uncertainties in this paper include: Ben Eaton – isotropic, anisotropic, dynamic and modified with correction factor; default from auto log calibration; Vernik, Jaeger & Cook; Hubbert & Willis; Thiercelin – MC envelope and stiffness tensors (Cij); Segall & Penebaker. The geomechanical properties from the different stress models noted above were inserted into a gridded fracturing simulator. The outputs were compared to actual job and calibration data for; minimum horizontal stress, end of job net pressure and fracture geometry for each of the models.
When comparing fracture geometries from each stress model against calibration data it is apparent that the chosen stress model will have a substantial influence on the result. This illustrates the importance of choosing the correct stress model for fracture simulations.
