Quantitative corrosion detection of reinforced concrete based on self-magnetic flux leakage and rust spot area

Abstract

The internal rebar corrosion of reinforced concrete (RC) structures harm the bearing capacity and durability of structures. Existing methods can measure rebar corrosion but are unsuitable for RC structures. Given this, the self-magnetic flux leakage (SMFL) field of V-shaped rebar corrosion damage was analyzed according to the magnetic dipole model, and the parameter K was proposed to characterize the corrosion degree. Using the naive Bayes algorithm, the SMFL method and the rust spot area analysis method were correlated to propose a rebar corrosion detection method. A corrosion detection experiment was conducted on RC specimens. The results showed that the parameter K was linearly correlated with the maximum cross-sectional rust loss rate η. Using the parameter K to evaluate the rust degree, the accuracy of rebar rust classification was 70%. After introducing the rust spot area ratio S as a supplementary parameter, the accuracy of rebar rust classification increased by 12.5% to 82.5%. This indicates that the proposed method could quantitatively detect the corrosion of the rebars within the concrete.