A Multi-Physics Coupling Driven Method for Initial Wellbore Trajectory Design

Wellbore trajectory design in complex geological formations is constrained by formation heterogeneity, structural discontinuities, and evolving in-situ stresses, making traditional geometry- or empiricism-based methods inadequate. This study introduces an innovative multi-physics coupling-driven framework that unifies geological modeling, drillstring statics, pore pressure–stress fields, and probabilistic risk mapping into a quantitatively constrained optimization system. The key innovation lies in coupling geo–mechanical stability analysis with Bayesian probability updating, enabling dynamic assessment of formation uncertainty and real-time trajectory feasibility. A B-spline-based parameterization combined with a hybrid intelligent optimizer ensures global convergence while balancing curvature smoothness and wellbore stability margin. Case results from a shale–sandstone formation in the Sichuan–Chongqing region demonstrate that the proposed approach reduces trajectory length by 7.1%, maximum dogleg severity by 35.4%, and high-risk interval penetration by 64%, with an 18.1% increase in reservoir exposure. This framework establishes a new paradigm for intelligent, uncertainty-aware trajectory design, effectively bridging formation physics, mechanical response, and probabilistic risk prediction.