Avaliação do risco de ruptura em análises de estabilidade de taludes de barragens de terra utilizando números fuzzy

• Ano de defesa: 2015
• Autor(a) principal: Silva, Amanda Vieira e
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Geotecnia; Lógica difusa; Barragens - Segurança
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/16526

Slope stability analyzes can be performed based on: (a) the limit equilibrium; and (b) the stress-strain behavior of the soil mass. The evalua tion of the risk of slope failure can be performed through probabilistic approaches, generat ing random values and assuming specific probability distributions for the geotechnical para meters of interest. This research aims to develop and implement a methodology for assessing t he risk of slope failure in earth dams, through applying the rules of operation with fuzzy numbers to the expressions for the factor of safety of Fellenius (1936) and Bishop Simplified (1 955) methods, as an alternative to probabilistic methods. The case of Olho d'Água Dam, built in the city of Várzea Alegre, State of Ceará, was chosen for the application and valida tion of the proposed methodology, considering operation with steady-state flow and ma ximum hydraulic load. Compared with the results, for the same dam, obtained by Araújo ( 2013), who assumed probability distributions, gamma for cohesion and beta for the friction angle, the fuzzy models, which employed trapezoidal and triangular membership func tions, maintained the conservatism trend, peculiar to the deterministic approaches of both used methods of slope stability analysis. The fuzzy model for Bishop Simplified (19 55) reported failure rates of 1% and 8% for the dam downstream slope, indicating a real ris k of failure (confirmed by its field performance, unsatisfactory since the start of oper ation), scenario not revealed by probabilistic analyses, which pointed out a zero pr obability of failure. The fuzzy methodology presented herein does not require the adoption of p robability density functions for the geotechnical parameters of interest, and provides t he realization of slope stability analyses within a much smaller computational time, without q uality losses in the results, requiring only the data of some laboratory tests, the technical ju dgment of an expert, and a spreadsheet. This makes the fuzzy approach a simpler alternative and faster than the probabilistic one, and with a promising future in its application to the analys es performed in everyday Geotechnical practice.