Objective Underground water-sealed oil and gas storage caverns are critical infrastructure for China’s national strategic energy reserves, requiring exceptionally stringent limits on cavern water inflow. At significant burial depths, surrounding rock masses are subjected to high in-situ stress, presenting a prominent engineering challenge in fractured rock: groundwater easily seeps through fractures, whereas grout struggles to penetrate tight fractures. Consequently, conventional grouting materials and construction techniques fail to satisfy the seepage-control requirements of compacted fractures under high-stress conditions. The engineering community still lacks a comprehensive understanding of the grouting and seepage-mitigation mechanisms within such fractured rock, as well as targeted technical routes and construction schemes. This gap remains a critical bottleneck restricting the safe and efficient construction of underground water-sealed storage caverns.

Methods To address these engineering challenges, field tests were conducted relying on a deep-buried underground water-sealed oil storage cavern project. The seepage characteristics and anti-seepage grouting mechanisms of the highly-stressed fractured rock masses were systematically analyzed to identify the fundamental causes of poor groutability. Based on these findings, three typical cavern sections exhibiting high-stress water seepage were selected, and comparative field pre-grouting tests utilizing Ordinary Portland Cement (OPC) and a novel ultra-fine cement material were performed. Three core evaluation metrics—unit cement consumption, inspection borehole water inflow, and post-excavation seepage performance—were adopted to quantitatively compare the seepage-control effectiveness and engineering adaptability of different grouting schemes.

Results OPC exhibited extremely poor groutability in the highly-stressed fractured rock. Its unit grout consumption remained consistently low without a distinct downward trend across sequential grouting cycles, resulting in a limited grout propagation radius. Even after multi-round sequential grouting, the maximum water inflow of the inspection boreholes far exceeded specified acceptance standards. By contrast, the ultra-fine cement significantly enhanced rock groutability, yielding markedly elevated unit grout consumption and a distinct decreasing trend of sequential grout consumption. Consequently, test sections grouted with ultra-fine cement exhibited far fewer seepage points and lower post-excavation water inflow than the OPC-grouted sections. Nevertheless, optimizing the grouting material alone was insufficient to achieve ideal seepage-control outcomes; inadequate grouting pressure failed to overcome the high in-situ stress to dilate compacted fractures, emerging as the dominant factor restricting target grouting performance.

Conclusion Highly-stressed fractured rock masses require a core technical route focused on expansion-type compaction grouting. To achieve this, an engineering implementation principle is proposed to accomplish comprehensive seepage control via one-off pre-grouting. These research findings provide crucial technical support for the seepage-control grouting design and construction of similar high-stress underground engineering projects.