Abstract: Aiming at the site boundary effects on the hypergravity shaking table, with the benefits of a newly developed flexible laminar container, we design and carry out comparison tests between the dry sand and the water centrifugal model. We introduce the metrics including the acceleration peak error EPGA, the normalized mean square error NMSE, and the peak frequency error EF, to evaluate the shear efficiency and boundary effects of the laminar container. The results show that compared with the center accelerations of the soil, the maximum values of EPGA, NMSE and EF for the laminar beams can be 21.50%, 21.36% and 3.1%, respectively, with the dry sand model, indicating the shear efficiency of the laminar container. Under the same dynamic load conditions, the peak acceleration and the peak frequency of each laminar beam in the dry sand model are 2.37 and 2.23 times of those of the water model, indicating that the response of the laminar beams in the dry sand model is controlled by the response of the soil, which further verifies the excellent shear efficiency of the laminar container. The shear modulus and the damping ratio are similar between the laminar beams and the center of the soil in the dry sand model, with the relative differences of 3% and 18%, respectively; the shear modulus of laminar beams in the water model is 1/3 of that in the dry sand model, which is only 9.22 MPa; but the damping ratio is relatively large, which is 34.8%. Taking 5% of the acceleration peak error as the threshold of the boundary effect, the test results are valid and reliable in the range of about 90 cm×26 cm at the center of the laminar container. The proposed methods and conclusions should have important guiding significance and application value for evaluating the boundary effect of the laminar container of the hypergravity shaking table.

Key words: hypergravity shaking table, flexible laminar container, boundary effect, evaluation indexes, verification methods