Análise de um caso de obra com contenções em estacas secantes

Abstract

Urban excavation of large basements close to neighboring buildings requires careful planning and construction operations to ensure small movements and no compromise to the integrity of the nearby constructions. Analyzing the pressures on retaining structures, sizing them appropriately, and predicting displacements adjacent to neighboring buildings are necessary steps. Each stage of the excavation must be monitored, as the geotechnical profile and conditions of nearby buildings can vary significantly. The use of numerical modeling software has proven to be quite efficient in aiding the development of such geotechnical projects, capable, in some cases, of reducing uncertainties inherent in traditional calculation methods. As technology evolves, different types of analyses have emerged, yielding increasingly realistic results. This research work aims to conduct an analysis of an instrumented case study located in the Jardim Botânico neighborhood in Rio de Janeiro, employing simplified analytical and numerical procedures using Finesoftware's GEO5 software. A 2D analysis was adopted, with parameter values obtained through correlations with field tests. Through the analyses conducted, it was possible to study the influence of parameters on the results via a sensitivity analysis. Additionally, by using horizontally displacement versus depth curves obtained numerically, the approach proposed by Ou et al. (1996) was employed, establishing a preliminary relationship between numerically estimated displacements in the plane strain condition and considering the three-dimensional effect, comparing the calculated displacements with those obtained experimentally using inclinometers. The displacements of the basement wall were instrumented, and all sample collection, characterization, special tests, and instrumentation activities are being conducted as part of a doctoral research project at UERJ. The results obtained in this study were consistent with expectations; even when using empirical correlations, the displacements obtained through analysis using GEO5 software exhibited behavior and magnitudes similar to those measured on-site. Horizontal displacements measured mostly fell within the numerically estimated results plus the correction method proposed by Ou et al. (1996) and the curves containing numerical results without the correction proposed by the authors. The measured horizontal displacements, in most cases, were found to be within the numerically estimated results, supplemented by the method proposed by Ou et al. (1996) and the curves containing numerical results without the correction proposed by the aforementioned authors.