Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
First-Order Reliability Method for Probabilistic Evaluation of Liquefaction Potential of Soil Using Genetic Programming
In this paper, liquefaction-triggering potential of soil is evaluated in terms of probability of liquefaction () using a first-order reliability method (FORM). First, an empirical model for determining the cyclic resistance ratio (CRR) of the soil was developed using multigene genetic programming (MGGP), an evolutionary artificial intelligence technique, based on the postliquefaction cone penetration test (CPT) data. This developed resistance model, along with the existing cyclic stress ratio (CSR) model, formed a limit-state function for a reliability-based approach to liquefaction-triggering analysis. The model uncertainty of the developed limit-state function was determined through an extensive reliability analysis following a trial-and-error approach, using Bayesian mapping functions that were calibrated with actual liquefaction field-performance observations of a high-quality, postliquefaction case-history database. A deterministic model with a mapping function relating and factor of safety against liquefaction () also was developed for use in the absence of parameter uncertainties. An example is presented to compare the present MGGP-based reliability method with the available ANN-based reliability method.
First-Order Reliability Method for Probabilistic Evaluation of Liquefaction Potential of Soil Using Genetic Programming
In this paper, liquefaction-triggering potential of soil is evaluated in terms of probability of liquefaction () using a first-order reliability method (FORM). First, an empirical model for determining the cyclic resistance ratio (CRR) of the soil was developed using multigene genetic programming (MGGP), an evolutionary artificial intelligence technique, based on the postliquefaction cone penetration test (CPT) data. This developed resistance model, along with the existing cyclic stress ratio (CSR) model, formed a limit-state function for a reliability-based approach to liquefaction-triggering analysis. The model uncertainty of the developed limit-state function was determined through an extensive reliability analysis following a trial-and-error approach, using Bayesian mapping functions that were calibrated with actual liquefaction field-performance observations of a high-quality, postliquefaction case-history database. A deterministic model with a mapping function relating and factor of safety against liquefaction () also was developed for use in the absence of parameter uncertainties. An example is presented to compare the present MGGP-based reliability method with the available ANN-based reliability method.
First-Order Reliability Method for Probabilistic Evaluation of Liquefaction Potential of Soil Using Genetic Programming
Muduli, Pradyut Kumar (Autor:in) / Das, Sarat Kumar (Autor:in)
20.11.2013
Aufsatz (Zeitschrift)
Elektronische Ressource
Unbekannt
| Taylor & Francis Verlag | 2014
First-Order Reliability Method for Probabilistic Liquefaction Triggering Analysis Using CPT
| British Library Online Contents | 2006
First-Order Reliability Method for Probabilistic Liquefaction Triggering Analysis Using CPT
| Online Contents | 2006
Probabilistic Evaluation of Liquefaction Potential
| British Library Conference Proceedings