Contribuição à confiabilidade de estacas cravadas através da interpretação de provas de carga dinâmicas com aplicação da teoria bayesiana

Abstract

Geotechnical works are one of the greatest examples of interaction between nature and engineering. Foundations accurately describe human s needs for predicting soil behavior, so the design of these structures requires specific considerations and actions in order to achieve safety. Most of the uncertainties inherent the foundation s project is related to the use of deterministic approaches in the designs and evaluations. Further, many inaccuracies originated from SPT tests contribute to increase deficiency in performance predictions of these structures. This aspect has led to the growing need for the development of methodologies that are able to reduce the intrinsic uncertainties of the behavior of foundations projects. In order to contribute to the study of the reliability of precast concrete piles, this dissertation proposes the analysis of a few piles executed in the district of Sepetiba, in the West Zone of Rio de Janeiro, tested through dynamic control methods. Moreover, in order to verify the peculiarities of three specific empirical methods, based on SPT results and assumed as prior distributions, the concepts of Bayesian analysis are used to estimate the subsequent updates of the pile load capacity, adopting as a likelihood function the results of the dynamic load tests interpreted by the CAPWAP computational tool. With the possibility of evaluating lateral friction and toe resistance separately, the same updates are done for both load capacity portions. By using the methodology initially proposed by Vrouwenvelder (1992), the study provides an application for the analysis of dynamic load tests, therefore, further studies of the reliability of driving piles with load tests results can be carried out. The subsequent calibration of the mathematical models parameters, used as a prior estimation, suggests a practical value of great relevance for technical and financial optimization of the projects and foundations works.