On the use of stochastic spectral methods in deep excavation inverse problems

Cañavate-Grimal, Antonio; Falcó, Antonio; Calderón, Pedro A.; Payá-Zaforteza, Ignacio

doi:10.1016/j.compstruc.2015.06.009

Cited by 5 publications

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The commonly used approaches to stochastic analysis in geotechnics are the Monte-Carlo method [2,10,11] and Latin hypercube sampling [12]. Another approaches exploit random sets [13] or crate meta models based either on Bayesian networks [14,15] or on the polynomial chaos expansions [16,17].…”

Section: Introductionmentioning

confidence: 99%

Uncertainty of material parameters for clay and its influence on simulation results

Janda

Proceedings of the Seventeenth International Conference On Civil, Structural and Environmental Engineering Computing

View full text Add to dashboard Cite

The paper presents how the uncertainty of material parameters of fine grained soils propagate to the results of numerical simulation. In particular, the posterior distribution describing the uncertainty of the five basic parameters of the hypoplastic model for clay is considered in two scenarios. In the first scenario, only the marginal distributions material parameters are considered independently and no correlation between them are assumed. In the second approach the complete multivariate posterior distribution is assumed in which the parameters are mutually correlated. This posterior distribution shows relatively low correlation between most of the parameter pairs with exception to the slope of primary consolidation line φ and its intercept N . The significantly higher variance of the simulation results obtained for the uncorrelated marginal distribution show that the correlations within the posterior distribution need to be considered in stochastic geotechnical simulations.

show abstract