This paper describes a method to compute the mobilised shear modulus, G, and damping ratio, D, using the non-linear fit of experimental transfer functions obtained at different depths in centrifuge models with the analytical expression of the amplification function for a visco-elastic soil layer on a rigid base. The corresponding shear strain, , is computed as a function of the particle velocity and shear wave velocity. The sources of potential error in the determination of G, D, and embedded in the proposed method are identified and discussed in comparison with two other methods that have been proposed in the literature, based either on the determination of the time lag of accelerations between two accelerometers or on the evaluation of the shear stressshear strain cycles from acceleration time histories recorded at different depths in the model. The performance of the three methods is evaluated using the experimental data obtained from nine centrifuge tests on dry sand. The values of G obtained by the proposed method compare very well with the results of laboratory and literature data; D values are more dispersed and slightly above literature data.
Conti, R., Viggiani, G. (2012). On the evaluation of soil dynamic properties in centrifuge tests. JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING, 138(7), 850-859 [10.1061/(ASCE)GT.1943-5606.0000659].
On the evaluation of soil dynamic properties in centrifuge tests
Conti, R;VIGGIANI, GIULIA
2012-01-01
Abstract
This paper describes a method to compute the mobilised shear modulus, G, and damping ratio, D, using the non-linear fit of experimental transfer functions obtained at different depths in centrifuge models with the analytical expression of the amplification function for a visco-elastic soil layer on a rigid base. The corresponding shear strain, , is computed as a function of the particle velocity and shear wave velocity. The sources of potential error in the determination of G, D, and embedded in the proposed method are identified and discussed in comparison with two other methods that have been proposed in the literature, based either on the determination of the time lag of accelerations between two accelerometers or on the evaluation of the shear stressshear strain cycles from acceleration time histories recorded at different depths in the model. The performance of the three methods is evaluated using the experimental data obtained from nine centrifuge tests on dry sand. The values of G obtained by the proposed method compare very well with the results of laboratory and literature data; D values are more dispersed and slightly above literature data.Questo articolo è pubblicato sotto una Licenza Licenza Creative Commons


