Interpretation of Rayleigh Frequency Shift Based Distributed Strain Sensing Data During Production and Shut-In of Unconventional Reservoirs

Abstract

Abstract A new fiber optic measurement, Distributed Strain Sensing based on Rayleigh Frequency Shift, (DSS-RFS) was recently applied to wells in unconventional reservoirs. During production operations, strain changes are measured with high spatial resolution and sensitivity. The objective of this paper is to show that it is possible and very useful to simulate and interpret these strain change plots to identify locations of producing perforation clusters and gain new insights into near wellbore fracture geometry. DSS-RFS fiber optic data is modeled during production from an unconventional reservoir. A fully coupled geomechanical fracture-reservoir simulator is used to simulate full lifecycle of hydraulic fractured horizontal wells, which incorporates the creation of hydraulic fractures with proppant injection, post-frac closure, primary production, cycles of production, shut-in and reopening. An implicit contact force model is implemented for modeling proppant embedment regarding fracture width change during fracture closure and pressure depletion. DSS-RFS plots are generated by obtaining the strain change along the wellbore during production. The simulations are then used to interpret the measurements in terms of pore pressure depletion and near wellbore fracture geometry. The simulated results match well with DSS-RFS data measured in the field. The tensional strain change signals correspond to the locations of active clusters, and the effect of pressure depletion is consistently seen in the simulations. This allows us to quantitatively interpret the measured DSS strain change in terms of the extent of pore pressure depletion during production. The strain change is also found to be related to near wellbore fracture geometry: (1) peak values of the tensional signals are positively correlated with near wellbore fracture width; (2) a larger simulated reservoir volume around a fracture leads to a wider positive strain change signal; (3) the height of the transition zone between active clusters is strongly related to reservoir depletion with respect to both space and time; (4) the height of the extensional signals can be used to assess production allocation among clusters when proppant injection distribution is relatively uniform during fracturing; (5) the shape the extensional signal becomes non-symmetric when there is a large depletion contrast on two sides of a fracture. A series of parameter sensitivity simulation results are analyzed to provide a systematic algorithm for accessing cluster efficiency and production allocation based on DSS-RFS data. The paper presents, a quantitative analysis for assessing cluster efficiency, location of active and inactive clusters and the extent of pore pressure depletion in a horizontal well using DSS-RFS strain change data. In addition, information about near wellbore fracture geometry can be inferred. This is made possible by a new and unique modeling capability that models the entire lifecycle of crack propagation with realistic fracture width and proppant volume distribution (considering stress shadow effects) as well as fluid production and well shut-in.