Distributed fiber optic strain measurements have become a widely used method for monitoring and diagnosing far-field strain induced during hydraulic fracturing. However, field observations have revealed that the mechanical decoupling between the fiber and the rock often happens. Despite this phenomenon, much of the current research assumes that the fiber cable remains mechanically coupled to the rock formation. To improve the accuracy and reliability of field data interpretation, it is essential to better understand the effect of fiber decoupling on strain responses. The objective of this study is to simulate strain responses while taking into account the impact of fiber slippage.
In this study, we developed a numerical model that allows for accurate prediction and analysis of strain responses of cross-well distributed fiber optic experiencing mechanical decoupling. Our model first uses the three-dimensional Displacement Discontinuity Method (DDM) to calculate displacement fields induced by hydraulic fractures. The model then takes into account the occurrence of slippage and calculate the fiber displacement distribution with slippage effect. Additionally, the model can predict fiber strain distribution in cases where the stick-slip behavior is present. To better understand the impact of friction on strain and strain-rate signals, we conduct a sensitivity analysis. This allows us to explore the effects of different friction levels on the signals, which can be critical in achieving more accurate and reliable results.
Our model results reveal several key observations. Firstly, when fiber slippage is detected, we typically observed a broadening of the extension zone near the fracture on the strain/strain rate waterfall plots, which cannot provide a well-define facture domain corridor. The simulated strain rate pattern matches the field observation. Additionally, the size of the decoupling zone increases as the friction between the fiber and formation decreases. Also, the decoupling generates negative cumulative strain change after a certain period of closure. Furthermore, when the dynamic friction is smaller than the static friction, sudden strain releases appearing as sticking portion were visible in the strain/strain-rate waterfall plots, indicating the occurrence of stick-slip behavior in the fiber. We also observed that the sudden strain releases in the strain rate waterfall plot occurs more frequent when the maximum fiber strain gradient is smaller. Moreover, the longer sticking portion was observed when the dynamic/static friction ratio is smaller.
The key innovation of this study is the development of a forward numerical model that simulates strain responses while taking into account the coupling effect between fiber and formation. To our knowledge, this is the first time that such a model has been presented. Overall, our study provides an important contribution to the field of hydraulic fracturing monitoring by enabling more accurate and reliable interpretation of strain signals, which can ultimately improve the safety and effectiveness of hydraulic fracturing operations.
