Prediction of lost circulation risk in fractured formations based on 3D geomechanical modeling

Due to complex geological structures and a narrow safe mud density window, offshore fractured formations frequently encounter severe lost circulation (LC) during drilling, significantly hindering oil and gas exploration and development. Predicting LC risks enables the targeted implementation of mitigation strategies, thereby reducing the frequency of such incidents. To address the limitations of existing 3D geomechanical modeling in predicting LC, such as arbitrary factor selection, subjective weight assignment, and the inability to achieve pre-drilling prediction along the entire well section, an improved prediction method is proposed. This method integrates multi-source data and incorporates three LC-related sensitivity factors: fracture characteristics, rock brittleness, and in-situ stress conditions. A quantitative risk assessment model for LC is developed by combining the subjective analytic hierarchy process with the objective entropy weight method (EWM) to assign weights. Subsequently, 3D geomechanical modeling is applied to identify regional risk zones, enabling digital visualization for pre-drilling risk prediction. The developed 3D LC risk prediction model was validated using actual LC incidents from drilled wells. Results were generally consistent with field-identified LC zones, with an average relative error of 19.08%, confirming its reliability. This method provides practical guidance for mitigating potential LC risks and optimizing drilling program designs in fractured formations.