The capability to correctly predict the liquefaction occurrence in numerical modelling has been largely assessed in many research groups and by adopting several constitutive models. However, less attention has been devoted to verifying the available approaches to simulate the soil behaviour from the small strain range to the liquefaction attainment. In this context, earthquake records provided by vertical liquefaction arrays constitute a powerful tool to verify the goodness of constitutive models because data associated to earthquakes of different level of intensity are available and can be used to explore a wide range of shear strains in the prediction of the dynamic nonlinear soil behaviour. This study presents the simulation of two well-characterized liquefaction arrays, located in Japan and California (US), respectively. The numerical analyses have been carried out by adopting a critical state compatible, stress ratio-based, bounding surface plasticity constitutive model, in the last released version, implemented in a commercial finite difference platform. The model calibration is based on the results of site investigations, with special attention to cyclic and dynamic laboratory tests collected in previous studies and not often considered in the calibration of the most recent advanced constitutive models. The simulation results are compared with the records at the surface and those obtained through an equivalent liner method. This study enriches the discussion on the calibration of soil models and contribute to a step forward in improving the reproducibility of the nonlinear behaviour of soils from the small-strain range until failure and after liquefaction.
Simulation of the seismic response of vertical instrumented arrays in liquefiable sites
A. Chiaradonna;P. Monaco
2024-01-01
Abstract
The capability to correctly predict the liquefaction occurrence in numerical modelling has been largely assessed in many research groups and by adopting several constitutive models. However, less attention has been devoted to verifying the available approaches to simulate the soil behaviour from the small strain range to the liquefaction attainment. In this context, earthquake records provided by vertical liquefaction arrays constitute a powerful tool to verify the goodness of constitutive models because data associated to earthquakes of different level of intensity are available and can be used to explore a wide range of shear strains in the prediction of the dynamic nonlinear soil behaviour. This study presents the simulation of two well-characterized liquefaction arrays, located in Japan and California (US), respectively. The numerical analyses have been carried out by adopting a critical state compatible, stress ratio-based, bounding surface plasticity constitutive model, in the last released version, implemented in a commercial finite difference platform. The model calibration is based on the results of site investigations, with special attention to cyclic and dynamic laboratory tests collected in previous studies and not often considered in the calibration of the most recent advanced constitutive models. The simulation results are compared with the records at the surface and those obtained through an equivalent liner method. This study enriches the discussion on the calibration of soil models and contribute to a step forward in improving the reproducibility of the nonlinear behaviour of soils from the small-strain range until failure and after liquefaction.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.