Reinforcement Detection in Concrete using Artificial Neural Network

In the diagnosis of reinforced concrete, investigating the size and cover thickness of the reinforcement is important. Electromagnetic radar, which conducts surveys by identifying scattered waveforms, is highly valued for its efficiency and convenience in these investigations. Recently, the use of Artificial Neural Networks (ANN) for identifying electromagnetic waveforms has been reported. While these reports mainly focus on void detection within concrete, the potential and accuracy of reinforcement detection using ANN are not examined in detail. In this paper, we explore the possibilities of estimating reinforcement size and cover thickness using ANN in a model of two-dimensional homogenous reinforced concrete. Furthermore, this paper reports on investigations conducted in cases where noise is introduced into the observed data and when multiple reinforcements are present.