Study on the influence of in-situ stress changes on shale fracture propagation considering the effect of effective stress coefficients

 [Objective] The geometry of multi-stage hydraulic fractures in shale gas horizontal wells is influenced by the three-dimensional in-situ stress distribution. The single-well stress profile serves as a crucial basis for predicting fracture height, and both the effective stress coefficient (Biot's coefficient) and the vertical grid resolution significantly impact the interpreted stress profile, consequently affecting the accuracy of fracture height prediction. [Methods] This study employs different Biot's coefficients (constant values and functions varying with logs) and vertical grid resolutions to compute stress profiles. It simulates and analyzes the differences in planar 3D fracture geometries under various stress profiles, systematically investigating the influence of Biot's coefficient on the stress profile and fracture geometry, and subsequently optimizing the vertical grid resolution and the method for interpreting Biot's coefficient. [Results] The results indicate that as Biot's coefficient decreases, the variation amplitude of the vertical in-situ stress profile increases, thereby restricting the vertical propagation capability of hydraulic fractures. Increasing the vertical grid resolution in the fracturing model helps to reduce the influence range of stress extremes; a 1 m vertical grid resolution achieves a favorable balance between simulation efficiency and accuracy. The Biot's coefficients calculated using empirical formulas and the poroelastic model yield similar results, with errors relative to laboratory measurements ranging from 3.68% to 3.93%. These methods provide a better match to stress test results from different formations. Furthermore, the simulated fracture heights using these variable coefficients align more closely with actual downhole fracture height monitoring results compared to using a constant Biot's value, showing errors of 8.64% to 9.94% compared to microseismic monitoring results from a vertical-to-horizontal well at the same site. [Conclusion] Through the analysis of initial in-situ stress fitting accuracy and the subsequent correspondence between simulated fracture height and monitoring data, it can be concluded that selecting an appropriate Biot's coefficient enables more realistic predictions of in-situ stress distribution and fracture propagation geometry.[Significance] This study provides valuable insights for future stress distribution calculations and fracture height predictions in shale gas wells through an in-depth discussion on the effects of Biot's coefficient and vertical grid resolution.