To accurately quantify pore pressure uncertainty and associated kick risk, this paper proposes a dual-phase pre-drilling risk assessment framework based on Bayesian Long Short-Term Memory (BLSTM) networks, integrating formation pressure prediction with distribution interference analysis. First, the effects of two Bayesian layer optimization methods—Monte Carlo dropout and Bayes-by-Backprop—on deep learning networks were systematically evaluated. The optimized Bayes-by-Backprop-LSTM model was subsequently selected for uncertainty prediction of formation pore pressure. Finally, kick risk was quantified by analyzing the interference between predicted pressure distributions and the safety margin of designed drilling mud density. The BLSTM models uncertainty regression between well-log parameters and formation pore pressure labels. Using the Bayes-by-Backprop strategy, it generates probabilistic pressure predictions. By incorporating the designed drilling mud density of target wells, kick risk probability is calculated through distribution interference criteria, where the overlapping area between pore pressure distributions and mud density safety boundaries is mapped to risk probability. Validation experiments utilized five types of well-log parameters from three wells in EAST CHINA. Key results demonstrate: (1) The BLSTM regression model achieved a mean absolute error (MAE) of 0.037 on test wells, representing a 26.7% reduction compared to conventional LSTM, with the 95% confidence interval coverage reaching 69.6%. (2) In the 3893–4048 m interval of a test well, interference areas exceeding thresholds indicated 60% kick risk probability. Spatial correlation with actual kick events revealed risk points undetectable by conventional pore pressure prediction methods. This study establishes a comprehensive risk assessment paradigm encompassing pore pressure uncertainty regression prediction and probabilistic risk calculation, providing drilling engineering with a framework that combines physical interpretability and statistical reliability.

This study aims to integrate extension limit screening, engineering evaluation, and casing-shoe depth search into a decision-support method for offshore extended-reach wells. The method combines open-hole, mechanical, and hydraulic extension limits with fuzzy scoring, analytic hierarchy process (AHP) weights, and a genetic algorithm (GA). The overall well score is used as the scalar search objective, with geological setting windows treated as prerequisites for engineering acceptance. The retrospective dataset comprises 30 completed wells, of which ten extension limit checks and five representative search cases are presented. Nine of the ten completed depths displayed are below the predicted limits; one exceeds its predicted limit by 113 m (1.83%). For Well A1, the open-hole, mechanical, and hydraulic limits are 5680 m, 4500 m, and 6200 m, respectively, making the mechanical limit controlling. The A1 numerical candidate scores 0.8898, below the field program at 0.8956, so the field program remains preferred. For Well A4, the actual four-section program reduces the final drilled interval from 2513.43 m in a hypothetical three-section program to 458.62 m and raises the overall score from 0.8562 to 0.8783. These cases provide retrospective evidence of workflow feasibility, not full predictive validation or proof of global optimality. Application requires conservative engineering screening and subsequent review of geological, trajectory, cementing, tubular-load, and rig constraints.