Open AccessNeural network modeling of concrete bond strength to reinforcement
Author(s) -
V. P. Yartsev,
A. N. Nikolyukin,
Anastasia Korneeva
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/687/3/033011
Subject(s) - reinforcement , bond strength , rebar , bond , shearing (physics) , materials science , structural engineering , bearing capacity , reinforced concrete , artificial neural network , compressive strength , composite material , computer science , engineering , layer (electronics) , artificial intelligence , adhesive , finance , economics
All the loss of bond strength of concrete to reinforcement is the main reason for loss of bearing capacity of a reinforced concrete structure. That’s why it is necessary to study the changes of bond strength of concrete to reinforcing bar by the influence of various factors. In addition, the mechanical characteristics of concrete change due to external and technological impacts. Making up an analytical model using artificial neural networks (NN), which allows determining the final bond strength through the mean values of shearing stress is considered in the article. The object of research is concrete of different strength classes, reinforced with steel and fibre-reinforced plastic rebar. The subject of study is the change in the value of bond strength of concrete to reinforcement after alternating freezing and thawing. It was found that the value of adhesion is associated with the strength characteristics of concrete and the type of reinforcement used. Also, a two-layer NN with reverse signal propagation was developed, which accurately describes the value of bond strength of concrete to reinforcement.