Prediction of shear capacity of single anchors located near a concrete edge using neural networks

• Published in: Computers & structures 2005-11, Vol.83 (28), p.2495-2502
• Main Authors: Alqedra, M.A., Ashour, A.F.
• Format: Article
• Language: English
• Subjects: Anchors; Applied sciences; Buildings. Public works; Capacity; Computational techniques; Concrete; Exact sciences and technology; Fracture mechanics (crack, fatigue, damage...); Fundamental areas of phenomenology (including applications); Geotechnics; Mathematical methods in physics; Neural networks; Physics; Shear; Soil mechanics. Rocks mechanics; Solid mechanics; Structural and continuum mechanics
• ISSN: 0045-7949; 1879-2243
• DOI: 10.1016/j.compstruc.2005.03.019

A feed forward back-propagation neural network model for predicting the shear capacity of anchor bolts located near a concrete edge is proposed. In the developed neural network, the neurons of the input layer represent the anchor outside diameter, concrete compressive strength, anchor embedment depth and the edge distance from the anchor bolt to the edge of concrete in the direction of the shear force. One neuron is used in the output layer to represent the concrete shear capacity of the anchor bolts. A database of 205 experiments available from previous laboratory anchor tests was utilised to train, validate and test the developed neural network. Predictions of the concrete shear capacity of anchors using the trained neural network are in good agreement with experimental results and those calculated from the concrete capacity design method. A parametric study has been conducted using the trained network to study the importance of different influencing parameters on the concrete shear capacity of anchor bolts. It has been shown that the concrete edge distance in the direction of the applied load has the most significant effect on the concrete shear strength of anchors.