ABSTRACT
In case of reinforced concrete (RC) beams strengthened by means of externally bonded FRP composites to enhance its flexural capacity, several premature failure modes can be detected, namely concrete cover separation (CCS), intermediate-span crack-induced (IC) de-bonding, or end-interfacial (EI) de-bonding. The mechanism of concrete cover separation and several available prediction models are identified in this study. Moreover, this paper explore the potential of gene expression programming (GEP) in developing a new model to predict concrete cover separation by training the model on experimental results of 127 data point failed by concrete cover separation. A comparative study was also conducted to assess the performance of the new GEP model against several available analytical models. The results demonstrated that the new developed GEP model can be employed as an efficient tool in predicting concrete cover separation failure load more accurately than the available analytical models.
