Prediction model for three-dimensional surface subsidence of salt cavern storage with different shapes

Abstract

Previous prediction models have faced challenges in characterizing the three-dimensional (3D) surface subsidence of salt caverns with different shapes. In order to overcome these limitations, this study proposes a novel theoretical model based on stochastic medium theory. This model provides a theoretical formula for characterizing the 3D surface subsidence of salt caverns with different shapes (oblate spherical, spherical, and vertical ellipsoidal). By employing the Particle Swarm Optimisation (PSO) algorithm to address the theoretical model, it can generate a three-dimensional settlement surface and subsequently compare it with the numerical results. The results demonstrate a strong consistency between the theoretical and numerical results. By conducting the parameter sensitivity analysis, it becomes evident that the shape of salt caverns holds paramount significance and should not be disregarded in the study of 3D surface subsidence. As the main influence angle increases and the burial depth decreases, the maximum settlement value of the salt cavern continues to increase while the maximum influence radius continues to decrease. The maximum settlement value and maximum influence radius of salt caverns gradually increase with the increase of the span. This suggests that the proposed theoretical model can effectively and intuitively describe the 3D surface subsidence characteristics of salt caverns.