Avaliação das abordagens em termos de tensão total e efetiva em estudos de estabilidade de aterros sobre solos compressíveis

Abstract

Nowadays embankments on soft soils have become very common due to the increasing unavailability of soils with great resistance, especially in industrial and urban high concentration places. However, the stability evaluation of these embankments is conditioned by the choice of strength parameters, the type of analysis adopted and the model chosen. Studies carried out in the Sarapuí region reveal that the choice of compressible material parameters is variable and difficult to obtain according to the type of test, the methodology applied and the different conditions under which the tests are carried out. In addition, there are also questions of correction such as the one proposed by Ladd and Lambe (1964), and evaluation of the sensitivity of these parameters. When evaluating the type of analysis adopted, in terms of total or effective stresses, both generate different safety factor results, which causes ambiguity about the design criteria determination. These stability assessments require care in the adoption of geotechnical parameters, use of the tool to evaluate the result and knowledge of the calculation methodologies performed by each adopted resistance model. At this moment, increasingly commercial geotechnical analysis programs are available performing complex calculations and in which the input data of the parameters is simplified, being up to the designer to choose them. This present study performed the retroanalysis of the stability of the Experimental Embankment I, situated in the baixada fluminense of Rio de Janeiro near the margins of the Sarapuí River, through finite elements modeling (FEM). The analysis in total stresses provided safety factors close to the unit considering undrained strength gains due to dissipation of excess poropressure during construction. The geotechnical profile of subsoil was based on vane tests which disregarded ground-steel friction. The analyses in effective terms also justified the rupture for a 2,5 m embankment height and allowed comparing the estimates of pore pressure profiles with the results obtained in the field. This study indicated that the retroanalysis in terms of total and effective stresses provided results compatible with the embankment failure condition. It is important to emphasize the relevance of the analysis parameters choice in the reliability of any numerical analysis.