Laboratory tests on the effects of partial embedment on soil-structure interaction (embedment effect test on soil-structure interaction)

A series of Model Tests of Embedment Effect on Reactor Buildings has been carried out by the Nuclear Power Engineering Corporation (NUPEC), under the sponsorship of the Ministry of International Trade and Industry (MITI) of Japan. Reactor buildings in Japan are partially embedded in general. Therefore, it is important to know how partial embedment affects the vibration characteristics of reactor buildings relating to seismic safety. Laboratory tests were conducted using a ground model made of silicone rubber (Young's modulus 2.3x106 Pa, Poisson's ratio 0.484, Density 1.24x 103 kg/m3 , Damping ratio 0.01) and a foundation model made of aluminum shown to study the effects of embedment on soil-structure interaction with different backfill types. The ground model is a cylinder, 70 cm high and 300 cm in diameter, with pit where the parallelepiped foundation model with square plan of 30 cm x 30 cm and 18 cm high was placed. Four types of backfill of silicone rubber (Young's modulus 3.4x105 Pa, Poisson's ratio 0.490, Density 1.00x103 kg/m3 , Damping ratio 0.03) softer than that of ground model filled the pit to realize non-embedment, full-embedment, one-side embedment, and two-sides embedment. Excitation types varied from hammering excitation at two different heights for acquiring impedance functions to shaking table excitation for foundation input motion. The authors have discussed the effect of one-side embedment on soil-structure interaction and this paper follows focusing on two-sides embedment. Impedance functions and foundation input motion of two-sides embedded foundation are firstly compared with directions regarding to the backfill, i.e., perpendicular and parallel, then with other embedment types. Two-dimensional FEM analysis is employed to simulate those functions for two-sides embedment