To predict the tensile capacity of adhesive anchors, a multilayered feed-forward neural network trained with the back-propagation algorithm is constructed. The ANN model have 5 inputs, including the compressive streng...To predict the tensile capacity of adhesive anchors, a multilayered feed-forward neural network trained with the back-propagation algorithm is constructed. The ANN model have 5 inputs, including the compressive strength of concrete, tensile strength of concrete, anchor diameter, hole diameter, embedment of anchors, and ultimate load. The predictions obtained from the trained ANN show a good agreement with the experiments. Meanwhile, the predicted ultimate tensile capacity of anchors is close to the one calculated from the strength formula of the combined cone-bond failure model.展开更多
基金The National Natural Science Foundationof China (No50578025)
文摘To predict the tensile capacity of adhesive anchors, a multilayered feed-forward neural network trained with the back-propagation algorithm is constructed. The ANN model have 5 inputs, including the compressive strength of concrete, tensile strength of concrete, anchor diameter, hole diameter, embedment of anchors, and ultimate load. The predictions obtained from the trained ANN show a good agreement with the experiments. Meanwhile, the predicted ultimate tensile capacity of anchors is close to the one calculated from the strength formula of the combined cone-bond failure model.
