Constitutive model for soil amplification of ground shaking: Parameter calibration, comparisons, validation

A one-dimensional constitutive model, developed for the nonlinear ground response analysis of layered soil deposits, is calibrated and validated experimentally in this paper. The small number of parameters renders the model easily implementable, yet quite flexible in effectively reproducing almost any type of experimentally observed hysteretic soil behavior. In particular, the model generates realistic shear modulus and damping curves as functions of shear strain, as well as stress–strain hysteresis loops. The model is calibrated against three sets of widely-used published shear modulus and damping (G : γ and ξ : γ) curves and a library of parameter values is assembled to facilitate its use. The model, along with a developed explicit finite-difference code, NL-DYAS, for analyzing the wave propagation in layered hysteretic soil deposits, is tested against established constitutive models and numerical tools such as Cyclic1D [12] and SHAKE [42], and validated against experimental data from two centrifuge tests. Emphasis is given on the proper assessment of the Vs profile in the centrifuge tests, on the role of soil nonlinearity, and on comparisons of two inelastic codes (NL-DYAS and Cyclic1D) with equivalent linear (SHAKE) analysis.

► An inelastic model for soil amplification is proposed, calibrated and validated. ► Calibration based on published G: γ/ξ: γ curves; parameter value library assembled. ► The model is tested against established constitutive models and numerical tools. ► It is also validated against experimental data from two centrifuge tests. ► The proposed model proved capable of predicting sufficiently seismic soil response.