An Integrated Experimental–Numerical Framework for Distributed Fiber-Optic Monitoring of Creep-Induced Deformation and Micro-Leakage in Deep Salt-Cavern Gas Storage Wellbores

Abstract

Abstract Long-term operation of deep salt-cavern gas storage subjects the wellbore barrier system to high temperature, high pressure, cyclic injection–withdrawal loads, and time-dependent salt creep. These coupled processes may lead to casing deformation, cement-sheath damage, annular sealing degradation, and eventually micro-leakage. Conventional integrity evaluation methods are usually intermittent and point-based, making it difficult to capture early, weak, and spatially uncertain anomalies. This paper proposes an integrated experimental–numerical framework for distributed fiber-optic monitoring of deep salt-cavern gas storage wellbores. The framework combines salt-creep/wellbore-barrier numerical simulation, full-scale creep-deformation physical simulation, gas micro-leakage simulation, DAS/DTS signal processing, and a real-time warning software architecture. A numerical modeling workflow is proposed to link salt creep, casing–cement deformation, and monitoring-feature selection. Once calibrated using site-specific salt-rock parameters, the workflow can be used to identify vulnerable intervals and guide the interpretation of low-frequency DAS responses. In the creep experiment, armored and helical optical cables were deployed at tubing-side and casing-side positions, and local loads of 10, 30, and 50 kg were applied to a simulated wellbore. Low-frequency DAS responses generally increased with deformation severity; for the armored cable deployed outside the tubing, the accumulated low-frequency response showed a representative linear relationship with load, y = 17.226x + 626.49, with R² = 0.959. In the leakage experiment, a 14 m full-scale simulator containing five controllable valves was pressurized to 2 MPa, and leakage rates from 5 L/min down to 0.1 L/min were tested. DAS responses in the 0–1 Hz and 1–50 Hz bands were most stable for micro-leakage identification, and the minimum detectable leakage rate reached 0.1 L/min under the laboratory conditions. The leakage-rate calibration for tubing-side leakage monitored by the inner cable was y = 0.9831x + 46.849, with R² = 0.8923. DTS clearly recorded imposed heating but did not show obvious temperature anomalies during micro-leakage, indicating that DAS should be the primary early-leakage indicator while DTS serves as an auxiliary thermal-state measurement. The experimental results demonstrate that calibrated DAS features can support deformation assessment, micro-leakage detection, and leakage localization under laboratory conditions. The numerical component provides a proposed methodology for future site-specific risk mapping, cable-deployment optimization, and warning-threshold development.