Search

Worldwide ageing infrastructures which are vulnerable to seismic lateral loads and located in high seismicity regions have arrested the interest of many researchers to find alternative materials and techniques to strengthen in bending and shear, for example reinforced concrete (RC) beams, slabs, columns, etc. There are several strengthening/repair techniques and materials in literature. Although the method of strengthening concrete structures with fiber reinforced polymers (FRP) is a relatively new technique, it has existed for more than two decades. In this context, several confinement models have been developed for FRP-confined concrete for the prediction of stress–strain response and several researchers have developed various constitutive models to measure the increase in the axial strength of concrete due to the confinement effect of FRP laminates. In this study, RC columns wrapped with carbon FRP (CFRP) considering some existing confinement models in the literature have been investigated. Moreover, based on the experimental data set in the literature, a new artificial intelligence-based algorithm (a Mamdani-type fuzzy inference system) was implemented to model the strength enhancement of CFRP confined RC columns using fuzzy logic methodology. Fuzzy logic predicted results were compared with the outputs of a non-linear regression analysis-based exponential model derived in the scope of the present work. The best predictive performances of the models were assessed by means of various descriptive statistical indicators. The comparison of the proposed prognostic approach with existing empirical and experimental data exhibits a very good precision of the developed artificial intelligence-based model in predicting the lateral confinement coefficient in CFRP wrapped RC columns. [ABSTRACT FROM AUTHOR]

[Display omitted] • CFRP-wrapped R/S RC columns were simulated using soft-computing methodology. • GPR, SVM, and M5P were inter-compared in prediction of K s for the first time. • GPR/SVM-based kernels and (un)pruned M5P were used for the first time for K s. • GPR-PUKF model outperformed than SVM and M5P models with lower deviations. • The total thickness of CFRP was the most effective parameter for predicting the K s. In this study, various soft-computing models (Gaussian process regression (GPR) and support vector machines (SVM) based on the polynomial kernel function (PKF), Pearson VII universal kernel function (PUKF), and radial basis kernel function (RBKF), as well as pruned/unpruned M5P tree models) were simultaneously applied for the first time in prediction of the lateral confinement coefficient (K s) of CFRP-wrapped rectangular/square (R/S) RC columns, and their corresponding predictive successes were appraised statistically. For this aim, short side of the column section (b), long side of the column section (h), total thickness of CFRP (t), compressive strength of the unconfined concrete (f' c 0), and the elastic modulus of CFRP (E CFRP) were used as independent input variables whereas the K s was the output variable. Results indicated that the performance of the Pearson VII kernel function-based Gaussian process regression (GPR-PUKF) model was superior to other models for the training and testing stages. A sensitivity investigation showed that the total thickness of CFRP (t) was the most effective parameter for predicting the K s using GPR-PUKF-based model. Findings of the present computational assessment obviously revealed that the employed soft-computing strategy had the capability of accurately estimating the K s of R/S RC columns wrapped with CFRP. [ABSTRACT FROM AUTHOR]