Abstract:

The shale gas reserves in the Wufeng Formation-Longmaxi Formation of the Luzhou Block in southern Sichuan are substantial. Tectonic movements alter the ground stress, significantly impacting the exploration and development of shale gas. To optimize exploration areas for deep shale, methods such as seismic comprehensive data, ancient structural maps, and rock mechanics parameter testing have been employed. Additionally, neural network algorithms and geological mechanics modeling analysis have been used to invert the stress field of ancient geological structures across multiple stages within the study area and to predict the development of reservoir fractures influenced by stress. The research indicates that numerical simulation methods and neural network algorithms effectively invert the crustal stress field across multiple stages. Tectonic movements have altered the crustal stress, concentrating it in the stratigraphic anticline. Here, the core of the anticline, affected by strong tectonic activity, is fractured, gradually releasing stress. The ongoing multi-stage tectonic movements have facilitated changes in the stress of the reservoir rock, making the fracture zone conducive to fault formation with decreasing stress over time. Around the original faults, crack development is pronounced, leading to stress attenuation zones prone to numerous, short, small cracks. The current stress field, shaped by multiple tectonic periods, presents a complex distribution and irregular crack development, significantly influencing shale gas drilling and development. These findings offer valuable insights for the exploration and development of deep shale gas.

Key words: shale gas reservoirs, multiphase tectonic movement, inversion of stress field, fracture evolution, neural network algorithm