ABSTRACT:

Low-frequency Distributed Acoustic Sensing (DAS) is a cross-well measurement technique for interpreting fracture geometry during the fracturing process. It measures strain and strain rate of the fiber in the observation well that geomechanical responses of the rock can be obtained to evaluate fracture geometry. However, this geomechanical process is not fully understood from the theoretical modeling perspective. In this study, we present a planar 3D multi-fracture model to simulate the strain and strain rate evolution in the fiber during the fracturing treatment. A planar 3D boundary element method is formulated to solve rock deformation, which is fully coupled with fluid flow in the deformable fracture. The strain and strain-rate of the rock and fiber are calculated at each time-step during fracture growth. The stresses calculated by the model have been validated against analytical solutions of a plane-strain fracture. The results show that the rock strain/strain-rate changes from tensile/extending to compressive/compressing when fracture hits the fiber, while the fiber measured strain/strain-rate keeps tensile/extending and converges to a narrowed zone after fracture hits the fiber. This difference can be explained as fiber measures average strain by the difference of axial displacement over the gauge length, which is different from rock strain when its value is not continuous across the fracture hit position. In the case of fracture hit, a strain-rate reversal occurs after pumping end, while strain-rate does not change its sign after pumping ends. The delay of strain-rate reversal after pumping ends can also be observed by the model, which may reflect the continuing growth of HFs after shut-in.

1. Introduction

Distributed Acoustic Sensing (DAS) has become a powerful diagnostic method for evaluating stimulation design by recording data from a treatment well or an offset well (Webster et al. 2014). DAS recorded in the offset/observation well can help evaluate the fracturing process when geomechanical responses, such as strain and strain-rate, are obtained (Ugueto et al., 2019). These geomechanical responses can reflect the moving fracture fronts or fluid fronts for multiple clusters. As low-frequency (<0.05 Hz) filter is adopted to extract the strain-rate from the recorded DAS data, this cross-well measurement is nameed low-frequency DAS (LFDAS) (Jin and Roy 2017).

This content is only available via PDF.
You can access this article if you purchase or spend a download.