Application of Artificial Neural Networks for Predicting the Bearing Capacity of Shallow Foundations on Rock Masses

Abstract

Calculation of the bearing capacity of shallow foundations on rock masses is usually addressed either using empirical equations, analytical solutions, or numerical models. While the empirical laws are limited to the particular conditions and local geology of the data and the application of analytical solutions is complex and limited by its simplified assumptions, numerical models offer a reliable solution for the task but require more computational effort. This research presents an artificial neural network (ANN) solution to predict the bearing capacity due to general shear failure more simply and straightforwardly, obtained from FLAC numerical calculations based on the Hoek and Brown criterion, reproducing more realistic configurations than those offered by empirical or analytical solutions. The inputs included in the proposed ANN are rock type, uniaxial compressive strength, geological strength index, foundation width, dilatancy, bidimensional or axisymmetric problem, the roughness of the foundation-rock contact, and consideration or not of the self-weight of the rock mass. The predictions from the ANN model are in very good agreement with the numerical results, proving that it can be successfully employed to provide a very accurate assessment of the bearing capacity in a simpler and more accessible way than the existing methods.

Assessment of the Bearing Capacity of Bridge Foundation on Rock Masses

“…Over the years, several methods were used to study bearing capacity: limit equilibrium method [1,2], slip line method [3], limit analysis method [4,5], numerical methods of finite elements (FEM) or DLO [6][7][8], and artificial intelligence techniques [9].…”

Bearing Capacity of Footings on Rock Masses Using Flow Laws

“…For the analysis of the bearing capacity of foundations, several procedures have been developed: the limit equilibrium method [1,2], the characteristic line method [3], the limit analysis method [4,5], the kinematical approach (as a part of limit analyses) [6], the numerical method [7][8][9] and artificial intelligence techniques [10,11].…”

Finite Element Analysis of the Influence of Artificial Cementation on the Strength Parameters and Bearing Capacity of Sandy Soil under a Strip Footing

“…In the field of geotechnical engineering, Terzaghi and Meyerhof methods remain those most commonly used [20] [21], although recent advances in technology have encouraged the use of software programs based on the finite element method. The use of the finite element method to solve various geotechnical problems is extremely useful, as a number of previous studies evidenced [22] [23].…”