A load carrying capacity of the reinforced concrete (RC) member is degraded by the corrosion of reinforcing steel bars due to chloride ion ingress. A lot of researches on the effect of corrosion in the longitudinal tensile reinforcing steel bars on the load carrying behavior have been available up to now. Accurate and quantitative estimation of capacity, however, is often difficult, because of the non-uniformity of corrosion in the member. Thus, a relationship between the spatial distribution of corrosion in the reinforcement including its scatter and the flexural loading capacity of RC member with such distribution of corrosion should be clarified so that the flexural capacity of corroded RC member can be estimated accurately. On the other hand, in case of the practical RC member under the corrosive environment, it should be considered that the flexural capacity often have to be derived from not a large number of inspection data on cross sectional areas of corroded reinforcements. So, in this study, a flexural loading test was performed by using RC beam specimens with the corroded tensile reinforcements provided the distribution of sectional areas. An estimation method of the flexural capacity of corroded RC beam was also shown, considering the distribution and its scatter in sectional areas of corroded reinforcements under the limited inspection data. Furthermore, the estimation of the longitudinal distribution of the cross sectional area of corroded reinforcement was performed by the spatial interpolation using Kriging method. Test results showed the yield and maximum load capacity in the corroded RC beam decreased as the corrosion rate increased. The failure mode of rupture in the reinforcement was shown in the large corrosion. The proposed estimation method was able to lead the safe evaluation of those experimental flexural capacities, determining the appropriate longitudinal characteristic value of the cross sectional area of corroded reinforcement. The flexural capacity can be also safely calculated using the characteristic value of diameters estimated by the corrosion crack width on the surface of the concrete, while the ratio of the experimental flexural capacity to the estimated one decreased as the corrosion loss increased. The distribution of bar diameters in the corroded reinforcement was able to be roughly estimated by using Kriging method. However, it was suggested that the measurement points close to the minimum bar diameter should be included to estimate the flexural capacity on the safe side.
Prediction of the Postfire Flexural Capacity of RC Beam Using GA-BPNN Machine Learning
