Objective Underground water-sealed cavern storages represent a crucial method for petroleum storage. A comprehensive evaluation of their seepage control and water sealing is essential during the construction process. This evaluation typically involves predicting the evolution of the seepage field in water-sealed cavern storages through numerical simulations, followed by an assessment of their seepage control and water sealing based on the simulation results. For large-scale water-sealed cavern storages, an equivalent continuum model is commonly used for simulations. The accuracy of these results is significantly influenced by permeability zoning. However, previous studies have often neglected to consider the impact of water-conducting zones on seepage fields, resulting in rough and insufficient permeability zoning.

Methods To enable rational permeability zoning within simulation ranges, a specific method for identifying water-conducting zones is proposed. Initially, a 3D geological distribution was developed based on the structural characteristics of the rock mass revealed during on-site cavern excavation. The subsequent analysis focused on identifying water-conducting zones with high hydraulic conductivity, utilizing data such as cavern water inflow, borehole water inflow, and changes in groundwater levels. A detailed equivalent continuum model was then established, incorporating the identified water-conducting zones. Numerical simulations were performed to calculate the seepage field of a specific underground water-sealed cavern storage under various conditions: initial excavation, grouting, and the implementation of an artificial water curtain system. The results were integrated into a comprehensive evaluation of the water sealing and seepage control of the water-sealed cavern storage.

Results Following the excavation of the caverns, the water head decreased rapidly across various zones, particularly the water-conducting zones. Under the initial excavation conditions, an unwatering zone developed in the surrounding rock of the main cavern, making it impossible to achieve the necessary water sealing conditions. After grouting was applied for plugging, rock mass permeability decreased significantly, with overall water inflow due to cavern excavation dropping by 63.1％. Notably, the water inflow in the water-conducting zones dropped by 70.3％, indicating a more pronounced effect of grouting in zones of strong permeability. However, the unwatering zone persisted in the surrounding rock even after grouting, resulting in continued challenges in meeting the water sealing requirements. Following the implementation of a water curtain system under a pressure of 0.3 MPa, the minimum water level in the cavern storage zones rose to 28 meters above the floor level of the main cavern, successfully satisfying the water sealing requirements.

Conclusion The proposed method for identifying water-conducting zones offers a valuable reference. Establishing a detailed model that accounts for these water-conducting zones can enhance the accuracy of simulation results. These findings provide a scientific basis for subsequent grouting efforts aimed at plugging.