Abordagem probabilística aplicada ao estudo da variabilidade geotécnica dos solos

• Ano de defesa: 2020
• Autor(a) principal: Izabela Couto Campello
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA TRANSPORTES E GEOTECNIA; Programa de Pós-Graduação em Geotecnia e Transportes; UFMG
• Palavras-chave em Português: Caracterização geotécnica; Incertezas; Variabilidade geotécnica; Análise probabilística; Comprimento de correlação espacial; Mecânica do solo; Taludes (mecânica dos solos); Probabilidades
• Link de acesso: http://hdl.handle.net/1843/35656

Uncertainties associated with the performance of geotechnical structures are unavoidable, whether due to limited site investigation, inherent soil variability or limitations in calculation models. It is part of a geotechnical engineer’s role, therefore, to select the appropriate approach to deal with these uncertainties, including conservative design, observational method during construction phase or probabilistic methods that account for uncertainties in the design. Several papers have been published in the last few decades examining geotechnical parameters variability and their effect on slope stability analyses. Despite the global increase, publications addressing this topic in Brazil are still restricted. This research aimed to present a method for modelling the uncertainties and variability of geotechnical properties focused on a tailings storage facility in Brazil. The methodology included: compilation of data from an extensive site investigation campaign; statistical and probabilistic data treatment, with focus on the coefficient of variation (CoV), probability distributions, cross-correlation and spatial variability and; assessment of geotechnical parameters variability and its effect on three cross-sections of a hypothetical dam. Sensitivity analyses carried out for the case study indicated a significant difference in the probability of failure given the variation of CoV, cross-correlation between cohesion and friction angle and vertical spatial correlation length. Hence, it was demonstrated the relevance of these parameters in the probabilistic slope stability approach.