National Technical University of Athens (NTUA)
Uncertainty quantification in Geotechnical Engineering
Abstract
dc:descriptionIn Geotechnical Engineering in all physical problems investigated, the variability ofthe soil is very large and at the same time the variability of the soil response shouldbe reduced in order to provide safe, reliable and economic structures. To this context,the uncertainty of properties such as the geometry of the problem analyzed, the typeand the amount of the loading, and the material variables that form the constitutivestress-strain law like Young Modulus, Poisson Ratio, failure stresses influence theoutput variables of displacements and stresses-strains of the aforementioned problems.In this thesis the quantitative effect of the uncertainty of the material inputvariables such as the compressibility factor of the constitutive relation, the criticalstate line inclination and the permeability in the output displacements, stresses andfailure loads is assessed. The physical problems examined in the thesis are the consolidationof fully saturated clayey soils and the corresponding failure of the soildomain. For this scope, computational tools and aspects of the probability theoryare implemented. More specific the stochastic finite element method is adopted usingthe random variables and the random field theory for representing the inputvariables and their variability. The Monte Carlo simulations with different ways ofsampling provide the output data and their statistical moments in order to quantifythe corresponding variability. This approach has been implemented to the consolidation of clayey soils takinginto account a material yield stress law of Modified Cam Clay type. It is portrayedthat in porous problems the coefficient of variation (CV) of the output is alwayssmaller than the CV of the input. Also, the output distribution remains Gaussiandespite the non linear relation between input and output variables. Consequently,in porous consolidation problems, the maximum displacement of the soil can bepredicted with smaller uncertainty and thus the soil structure interaction design ismore accurate. The aforementioned framework is adopted to the failure of clayey soils, afterthe pore pressure is set to an equilibrium. It is confirmed that both failure loadand displacements follow Gaussian normal distribution despite the non linearity ofthe problem. Moreover, as the soil depth increases the mean value of failure loaddecreases and the failure displacement increases. Consequently, failure mechanismof clays can be determined in this work within an acceptable variability, taking intoaccount the soil depth and non linear constitutive relations which in the analyticalsolutions is not feasible as it is assumed the Meyerhoff theory which considers theelastic halfspace.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Savvides, Ambrosios-Antonios
- Σαββίδης, Αμβρόσιος-Αντώνιος
Subjects
dc:subject × 17- Στοχαστικά πεπερασμένα στοιχεία
- Σειρές Karhunen Loeve
- Συνεκτικά εδάφη
- Στερεοποίηση Αργιλικών Εδαφών
- Επιφανειακές θεμελιώσεις
- Αστοχία αργίλων
- Stochastic finite element method
- Karhunen Loeve Series
- Cohesive soils
- Porous Consolidation
- Shallow foundations
- Failure of clays
- Επιστήμη Πολιτικού Μηχανικού
- Επιστήμες Μηχανικού και Τεχνολογία
- Πολιτική μηχανική
- Civil Engineering
- Engineering and Technology
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/50006