There are many natural fractures in shale reservoirs, changes in hydraulic fracture extension patterns. In the paper, a multi-fracture extension finite element model for horizontal wells in shale reservoirs under the disturbance of natural fractures is established by combining the actual geological parameters and construction parameters of a horizontal well multi-fracturing operation in X oilfield to analyze the effects of the difference in geostress, elastic modulus, angle of natural fractures, and the number of natural fracture groups on the hydraulic fracture extension. The results show that with the increase in ground stress difference and natural fracture angle, hydraulic fractures are more likely to penetrate the natural fractures; with the increase in elastic modulus, the fracture stress and tip stress increase, the volume of rupture unit, the fracture extension width and the pore pressure concentration area decrease, and it is easy to form a long and narrow fracture; with the increase in the number of fracture groups, the connectivity of reservoir fractures increases, the extension of fractures is stronger, and it is easier to form a complex fracture network.

The pore degree prediction method based on well logging interpretation for tight sandstone reservoirs cannot meet the requirements of timeliness and rapidity for well exploration decisions. This paper utilizes the logging parameters during drilling, combined with acoustic time difference experiments, dynamic and static parameter experiments, and full-scale drill bit rock-breaking simulation, to reveal the response characteristics of reservoir properties to the feedback information from logging and rock-breaking and establishes a pore degree prediction method. The results show that as the pore degree decreases, the dynamic and static elastic modulus increases, the rate of penetration decreases, the torque increases, the mechanical specific energy increases, and a mathematical relationship model between pore degree and mechanical specific energy is established, achieving real-time drilling prediction of pore degree. The new method has been applied in the NB block, and the coincidence rate with the well-logging interpretation results reaches over 83%. The research results have provided real-time predictions of reservoir pore degrees and improved the efficiency of exploration decisions.