| Resumo: |
Cantilevered retaining walls are generally designed as continuous structures using limit equilibrium theories. However, these methodologies do not consider specific aspects such as the arching effect, which occurs in the soil when the structure is composed of closely spaced juxtaposed piles. Approaches that consider triple states of stresses and strain (3D analyses) allow for a detailed analysis of the stress-strain behavior of soil masses, capturing the complexity of stress redistribution associated with the arching effect. This is particularly relevant for retaining walls formed by juxtaposed elements, where the plane strain condition (2D analyses) is not valid. Thus, this work aims to present an initial proposal for estimating the spacing between piles of cantilevered retaining walls formed by juxtaposed concrete piles through results from numerical simulations of stress-strain behavior obtained by 3D analyses. The simulations were carried out using the finite element method, through the RS2 and RS3 software from Rocscience. The soil resistance parameters (cohesion and friction angle) and the configuration of the retaining wall were varied in the simulations. Proposals for estimating the spacing between piles were obtained using linear and nonlinear regression with the aid of Excel software. The results showed that the spacing between piles reveals a relief of stresses between the pile faces and an increase in stresses near them, characterizing the active arching of the soil. The increase in soil cohesion results in higher confining stresses, reducing displacements and soil plastification, indicating a greater tendency for arching formation. Significant differences in stress and displacement fieldswere observed between the 2D and 3D analyses of the retaining walls. The 2D analyses do not satisfactorily assess the influence of spacing between structural elements on the stress field. Conversely, the 3D analyses indicate that decreasing the spacing between piles increases the confining stresses, favoring the formation of the arching effect, especially in cohesive soils. Additionally, displacements in the 2D analyses were greater than in the 3D simulations, suggesting a more conservative approach. The analyses allowed for the definition of preliminary models to estimate the spacing between piles. It was found that soil cohesion and pile diameter positively influence the models, while excavation depth has a negative effect. It is recommended to adopt the confidence interval as a safety factor in the results obtained by the proposed equations. |
|---|
