Export file:


  • RIS(for EndNote,Reference Manager,ProCite)
  • BibTex
  • Text


  • Citation Only
  • Citation and Abstract

Modified regression and ANN model for load carrying capacity of corroded reinforced concrete beam

Department of Civil Engineering, SHUATS (Formerly AAI-DU), Allahabad-211007, U.P, India

Topical Section: Corrosion

There have been many extensive studies on the prediction of the residual strength of corroded reinforced concrete beams from experimental and theoretical perspectives in the past. This article corroborated the findings of Azad et al. (2010) pertaining to the residual strength and safety of the corroded beams and an insight to develop an improved regression model to obtain more practical outcomes. The proposed model has further been verified with the past research data to obtain a validation error to its minimum count. The study is also followed by the use of soft computing technique like Artificial Neural Networks (ANN) to establish a method with substantial improvement in the prediction results of residual strength. One ANN model is proposed to predict the residual capacity of corroded reinforced concrete beams using the same data from Azad et al. (2010). The effects of fixed data stratification on the performance of the models have been studied. The results of the ANN model were found to be in good agreement with experimental values. When compared with the results of Azad et al. (2010), the ANN model with fixed data stratification gave a better prediction for residual strength with reference to correlation coefficient and error reduction. Hence, the reliability of ANN model is assured with the prediction work followed in this study.
  Article Metrics
Download full text in PDF

Export Citation

Article outline

Copyright © AIMS Press All Rights Reserved