Geomechanics and Wellbore Stability

Eriksfiord applies geomechanics in two different ways:

  • Analytical methods and elastic theory provide fast solutions for simple geometries such as a round borehole, and allow us to predict wellbore stability in a known stress field, and inversely, to calculate rock stress from features observed on the borehole surface by borehole imaging (Vinland Software Suite). Thanks to speedy calculations, sensitivity to specific parameters is easily tested and quantified.

For anisotropic or time-dependent effects, we employ numerical methods.

  • When no borehole images are available, or when evidence is ambiguous, a numerical method may provide a solution. If an oilfield has a pronounced structural trend, a simple 2D simulation based on Finite Difference (grid) or Finite Elements (mesh) may demonstrate stress features which influence migration, drilling or production behaviour.
 

The first step of a routine geomechanical analysis is to construct a simplified 2D profile (based on a seismic section; for strike-slip, a horizontal model may be appropriate) followed by FEM solution for stress in a given tectonic situation.

The example below shows an area of low stress below a salt diapir (leading to mud loss risk) and a 45deg deflection of stress orientation near the diapir wall. A tectonic hypothesis can be tested, such as “what is the influence of basement topography and geodynamic stress on fracture pressure”.

 

In contrast to this generic approach aimed at demonstrating a hypothesis, numerical 2D analysis gives quantitative answers if realistic geometry, rock properties and pressures are entered. Stress and fluid flow can be coupled, e.g. to show how oil production affects stress distribution, or to elucidate migration pathways during tectonic deformation (tectonic pumping).

Where there is no clear structural trend, a 3D model may be necessary to explain borehole observations. A simplified version of the geometric model (from RMS, Petrel or similar systems) is transferred for meshing or gridding. The 3D model provides a stress tensor for every (x,y,z) in the model and can be coupled with a fluid flow simulation. In detailed realistic 3D models, observed results may be of great significance, but difficult to explain and to verify.

 

Cobra (Borehole Instability Log)

From the model, curves of stress and pressure can be extracted along a welltrace, and compared with well log observations. We call this method Cobra and an example is shown below.

Color-coded bands are displayed on a log plot track from 0 to 180 degrees of relative bearing. Each band corresponds to a sector of the borehole and shows either compressive or tensile instability, calculated by a combination of pore pressure, stress and rock properties at a certain depth.

Cobra is based on elastic theory (Stassi D’Alia, 3D Lade, Mogi-Coulomb and Mohr-Coulomb failure criteria). The prediction of rock strength is calibrated to test data through a multivariate procedure.