Evaluation of Clay Constitutive Models for Analysis of Deep Excavation under Undrained Conditions

Abstract

The objective of this study is to evaluate the performance of the most commonly used constitutive models of clay for analysis of deep excavation under the undrained condition. Five soil models, i.e., the modified Cam clay model, the hardening soil model, the hardening soil small strain model, the Mohr-Coulomb (φ = 0) and the undrained soft clay model, are selected for evaluation. The TNEC excavation case history, which was with well-documented monitoring data and soil properties, is used for this purpose. Results indicate that the modified Cam Clay model, with a raise of the κ/λ ratio to 0.2 ~ 0.25 for normally consolidated clay and without adjustment of κ/λ ratio for overconsolidated clay, can yield predicted wall deflections close to field measurements for the final stage of excavation. The hardening soil model and hardening soil small strain model with the parameters directly obtained from tests can yield wall deflections close to field measurements for the final stage of excavation. The φ = 0 Mohr-Coulomb model with Eu / su = 500 can result in good prediction for wall deflections for final stage of excavation. None of those four soil models can predict good ground settlement profiles. The undrained soft clay model with the parameters directly from tests can predict both wall deflections and surface settlements well.