Bearing capacity of strip footing resting above a circular void in the rock mass using adaptive finite element method

Piyush Kumar,Vinay Bhushan Chauhan

DOI: https://doi.org/10.1007/s41062-021-00666-y

2021-10-25

Innovative Infrastructure Solutions

Abstract:Strip footing needs special attention when substantial underground voids of significant dimensions occur in natural rocks since voids can compromise structural integrity. This paper uses a numerical simulation, based on finite element limit analysis with adaptive mesh refinement technique, to evaluate the effect of a circular void in the rock mass on the ultimate bearing capacity of a uniformly distributed loaded strip footing having a width, B, assuming that the rock mass obeys the generalized Hoek–Brown failure criterion. Using parametric studies, the effect of several governing factors on the ultimate bearing capacity of the strip footing is investigated. The parametric consideration involves the void diameter, the vertical distance to the footing width, and the horizontal void’s distance to footing width from the axis of footing, and the various rock parameters such as uniaxial compressive strength, unit weight, geological strength index, material constant, and disturbance factor. The effect of voids on the ultimate bearing capacity of footing is quantified and a reduction factor, Rf\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$R_{f}$$\end{document}, is proposed to set up the design charts from the numerical simulations. Moreover, the influence of the void’s presence on the development of potential failure planes below the footing is examined and discussed in detail. The results showed that the existence of a void beyond a depth of 3B, does not affect the footing’s ultimate bearing capacity, and a void at a horizontal distance of 0.5B was found to be a critical one. Additionally, a multiple linear regression model has been developed to predict the Rf\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$R_{f}$$\end{document} with good accuracy, as the proposed model’s coefficient of determination (R2) is noted to be 0.94.