A data mining approach to compressive strength of CFRP-confined concrete cylinders

Abstract

In this paper, compressive strength of carbon fiber reinforced polymer (CFRP) confined concrete cylinders is formulated using a hybrid method coupling genetic programming (GP) and simulated annealing (SA), called GP/SA, and a robust variant of GP, namely multi expression programming (MEP). Straightforward GP/SA and MEP-based prediction equations are derived for the compressive strength of CFRP-wrapped concrete cylinders. The models are constructed using two sets of predictor variables. The first set comprises diameter of concrete cylinder, unconfined concrete strength, tensile strength of CFRP laminate, and total thickness of CFRP layer. The most widely used parameters of unconfined concrete strength and ultimate confinement pressure are included in the second set. The models are developed based on the experimental results obtained from the literature. To verify the applicability of the proposed models, they are employed to estimate the compressive strength of parts of test results that were not included in the modeling process. A sensitivity analysis is carried out to determine the contributions of the parameters affecting the compressive strength. For more verification, a parametric study is carried out and the trends of the results are confirmed via some previous studies. The GP/SA and MEP models are able to predict the ultimate compressive strength with an acceptable level of accuracy. The proposed models perform superior than several CFRP confinement models found in the literature. The derived models are particularly valuable for pre-design purposes.