This paper investigates the effectiveness of four different soft computing methods, namely\nradial basis neural network (RBNN), adaptive neuro fuzzy inference system (ANFIS) with subtractive\nclustering (ANFIS-SC), ANFIS with fuzzy c-means clustering (ANFIS-FCM) and M5 model tree\n(M5Tree), for predicting the ultimate strength and strain of concrete cylinders confined with\nfiber-reinforced polymer (FRP) sheets. The models were compared according to the root mean\nsquare error (RMSE), mean absolute relative error (MARE) and determination coefficient (R2) criteria.\nSimilar accuracy was obtained by RBNN and ANFIS-FCM, and they provided better estimates\nin modeling ultimate strength of confined concrete. The ANFIS-SC, however, performed slightly\nbetter than the RBNN and ANFIS-FCM in estimating ultimate strain of confined concrete, and\nM5Tree provided the worst strength and strain estimates. Finally, the effects of strain ratio and the\nconfinement stiffness ratio on strength and strain were investigated, and the confinement stiffness\nratio was shown to be more effective.
Loading....