Abstract
In recent years, projects have been proposed to utilize salt caverns as a storage method for supercritical CO2 (s-CO2) and have been carried out around the world, which can effectively reduce the anthropogenic greenhouse gases (GHG) concentration. Careful and rational design of the salt cavern are required to guarantee the structural stability and reliability of the cavern when in service. In this study, physics-based FE models were first developed to simulate the salt cavern construction process and predict the creep deformation of the cavern wall during s-CO2 storage. In addition, Gaussian process (GP) based surrogate models were constructed from FE simulation results to estimate the properties of salt caverns with cheaper calculations. Then, to obtain a reliable and robust design, an RBDO framework was employed with the assistance of an adaptive fidelity enhancement technique to scan through the high-dimensional design space, which provides a swift and efficient search for the optimal conditions while considering the uncertainties in the salt cavern construction process. This research is one of the first to account for uncertainties within the salt cavern design process, and the results show that uncertainties of variables can strongly affect the system reliability and the volume of the salt cavern.
