Avaliação da estabilidade de barragens brasileiras para acumulação de água submetidas a carregamentos sísmicos: estudo de caso da barragem Armando Ribeiro Gonçalves

Abstract

Dams are complex structures, that experience, in normal conditions, constant loads over this service life. This stability condition can be modified by exceptional loadings, like earthquake loads. Brazil is in an intraplate position, with low levels of seismic risk, however, especially in some regions such as the region northeast, this seismic risk may be relevant. Therefore, it is necessary develop analyzes of seismic stability with dams located in these regions. In this way, Armando Ribeiro Gonçalves dam is one of the most important dams in the northeast region, being marked by relevant and documented seismic activity. Before the presented, dynamic analyzes were developed with Armando Ribeiro Gonçalves dam, by two different methods, aiming to understand the Armando Ribeiro Gonçalves dam response when subjected to different levels of seismic loading, using different methodologies of analysis. The dam was evaluated at three water levels in the reservoir. Stress x Strain and stability analyzes were developed in the static condition, to enable the development and serve as a reference value to the dynamic analyses. Pseudo-static analyzes were developed in a way, using different methods to define the acceleration coefficients. In addition, dynamic analyzes with the Linear Equivalent constitutive model and post-earthquake stability analyzes were also developed, using accelerograms defined by return times between 475 and 10,000 years. The pseudo-static analyzes results indicate stability, using the recommendations of ANA (2016), to cases where isn’t a local study, DSD (USBR, 1973), and the method of Cruz et al. (2021 and 2022), considering return times of 475 and 975. The analysis presents inadequate safety factor in the other situations. The dynamic analysis results indicate stability for the analysis with earthquakes defined by return times between 475 and 2,475 years, with ruptures being perceived for the earthquake defined by return time of 10,000 years, caused by pore pressures appearance. A Good agreement was observed between the results for different analysis methodologies, although, the pseudo-static analysis is unable to identify ruptures caused by pore pressure appearance. Finally, it was possible to perceive that the water level retained in the dam has a smaller effect on stability, mainly on the downstream face