The freezing method is an important construction method for crossing water-rich soft rock formations. The long-term stability of the frozen wall plays an important role in engineering safety. Creep damage is one of the remarkable features of induced deformation of frozen walls. It is of great theoretical and engineering significance to study the characteristics of frozen rock creep. Taking the cretaceous saturated frozen sandstone as the research object, triaxial creep mechanical tests with different confining pressures (0, 2, 4 and 6 MPa) were carried out under the low temperature freezing condition of -10 degrees C. The creep deformation of saturated frozen sandstone was analyzed. According to the existing viscoelastic-plastic model, the parameter identification was carried out and the variation law of creep parameters was studied. A creep constitutive model considering temperature and damage effect was then proposed. The results show that the low temperature freezing weakens the mutual cementation force between the particles during the creep process, and the creep characteristics are obvious. However, the confining pressure inhibits the development of the internal damage of the saturated frozen sandstone to some extent, resulting in the steady creep rate. The increase in confining pressure shows a significant downward trend. With the increase of confining pressure, the creep failure morphology of saturated frozen sandstone shows a change process from shear failure to tensile failure to local shape hardening failure. On the basis of the viscoelastic model, the creep parameters E-1, E-2 and eta(2) are increased first and then decreased with the increase of the load, and the inflection point is the yield stress. The parameter eta(0) appears after the yield stress and undergoes increase to decrease. Combined with the frozen rock creep data, the parameters of the stress and low temperature coupled creep constitutive model are identified, and the numerical results of the model are compared with the creep experimental data to verify the correctness and rationality of the established nonlinear model.
