Quantitative prediction of multistage fractures of ultra-deep tight sandstone based on the theory of minimum energy dissipation

Due to strong reservoir heterogeneity and low-resolution limit of geophysical data, it is difficult to predict fractures in ultra-deep reservoirs by conventional methods. In this research, we established a novel geomechanical model for prediction of fracture distribution in brittle reservoirs, especially for ultra-deep tight sandstone reservoirs. Methodologically, we intended to introduce the minimum energy dissipation principle considering time variable, combined with the generalized Hooke's law containing damage variable, and obtained the energy dissipation rate expression corresponding to the energy dissipation process of brittle rocks. Combined with the three-shear energy yield criterion, the Lagrangian multiplier was introduced to deduce and construct the constitutive model and the failure criterion of rocks under the framework of the theory of minimum energy dissipation. Based on the law of conservation of energy, the stress-energy coupling characterization model of fracture density parameter was derived. Finally, all the improved geomechanical equations were incorporated into a finite element software to quantitatively simulate the distributions of tectonic stress filed and fractures based on paleo-structure restoration of Keshen anticline during the middle and late Himalayan periods. Its predictions agreed well with measured fracture density from reservoir cores and image logs.