Yiyang Wang

With the rapid development of clean energy in China, the demand for underground energy storage is increasing. This paper reviews the current status of underground energy storage and advances in safety assessment and environmental-impact mitigation. Depleted oil and gas reservoirs and salt caverns are the primary geological formations for natural-gas and compressed-air energy storage in China. As of October 2024, eight salt-cavern and 26 depleted-reservoir facilities had been constructed, accounting for approximately 90% of large-scale underground storage facilities. Oil and natural gas are the most mature storage media, followed by compressed air and carbon dioxide (CO₂), whereas underground hydrogen and helium storage remain at earlier stages of development. The review compares salt caverns, depleted reservoirs, aquifers, and rock caverns in terms of capacity, sealing, cost, cycling capability, and risk. It further synthesizes theoretical, numerical, experimental, and monitoring studies. The principal risks include well or fracture leakage, caprock or cavern sealing loss, pressure-induced fracturing, salt creep, cyclic damage, hydrogen-related microbial and geochemical reactions, material degradation, and CO₂ plume migration. Long-term monitoring should combine pressure, temperature, flow rate, gas composition, well integrity, deformation, microseismicity, groundwater chemistry, surface-gas geochemistry, and plume imaging. Future development should advance abandoned-mine pumped-storage hydropower, CO₂ geological storage, and underground hydrogen storage through field-scale validation, coupled multi-physics models, hydrogen-compatible materials, and unified integrity standards.

Large-scale experiments and integral-model analysis were performed to examine horizontal freshwater–brine buoyant jets relevant to salt cavern leaching. Freshwater was discharged into initially quiescent saturated brine in a cylindrical tank with a diameter of 13 m and a height of 8 m, using nozzle diameters of 0.10–0.26 m and inlet velocities of 1.0–4.0 m/s. The tests covered nozzle Reynolds numbers of 1.57×105–6.27×105 and initial densimetric Froude numbers of Fr0=1.78–7.14. Increasing Fr0 enhanced horizontal penetration and delayed buoyant rise. The classical momentum–buoyancy length scale LM helped organize the measured centerline trajectories, whereas the measured characteristic half-width was better organized by the source length scale LQ, indicating that the two morphology measures are more effectively represented by different scales within the tested range. A steady centerline-marching integral model captured the mean centerline trend but showed stronger sensitivity in half-width prediction; model comparisons indicate an effective entrainment-coefficient range of α=0.075–0.090. Empirical correlations are fitted for the trajectory and half-width within the tested parameter range. Together, these results provide a large-scale dataset and a quantitative assessment of classical scaling and entrainment-based integral modeling for freshwater–brine jets with a strong density contrast relevant to horizontal salt cavern leaching.