Abstarct: The solution mining of salt has been in use for many years. Cavities formed by this method are used for storage of compressed air energy, natural gas, petroleum and many other types of hydrocarbons. For natural gas storage salt caverns a seasonal storage pattern has been applied so far. According to increasing demands of the international gas market, cavern storage. Operators are asking more frequent cycling of internal pressures at higher rates in terms of mass flow and pressure gradients with time. In order to guarantee a safe operation for high frequency cycling storage caverns, the rock mechanical layout and design concepts have to be reviewed and extended. The objective of current dissertation dedicated to the study of the mechanical behavior of salt caverns using experimental studies. To obtain the salt rock properties, static tests are designed and executed. To determine time-dependent properties of the salt rock, long-term creep testing equipment designed and manufactured baxsed on mechanical loading system using second-type lever arms. Also, cyclic loading test with mean load level greater than zero were conducted to simulate ratcheting and shake-down behavior of salt caverns with frequencies ranging from monthly to 0.25Hz. Moreover, different loading amplitudes were applied to investigate the impact of cavern depth on volume loss. In order to predict time-dependent properties of salt rock, LUBBY2 constitutive model was used. To promote the model performance, a new model was presented by using loading. Evaluating the new constitutive model, both static and dynamic step-wise creep tests were performed and compared with the model results. It can be concluded that the new model can effectively consider the effects of frequency on creeping behavior of salt rocks. Viscoplastic and viscoelastic parameters will exponentially decrease by increasing loading frequency. Also, loading amplitude showed an outstanding influence on creep parameters of salt rock. Numerical modeling conducted to predict the effects of shape, loading frequency and depth of the cavern on its convergence. The results reveal that higher height to diameter ratio; loading frequency and deeper cavern last casing shoe lead to higher convergence rate.