Using the Artificial Neural Network to Predict the Axial Strength and Strain of Concrete-filled Plastic Tube

Document Type: Regular Article


University of salahaddin-Erbil


The main purpose of the current study was to formulate an empirical expression for predicting the axial compression capacity and axial strain of concrete-filled plastic tubular specimens (CFPT) using the artificial neural network (ANN). A total of seventy-two experimental test data of CFPT and unconfined concrete were used for training, testing, and validating the ANN models. The ANN axial strength and strain predictions were compared with the experimental data and predictions from several existing strength models for fiber-reinforced polymer (FRP)-confined concrete. Five statistical indices were used to determine the performance of all models considered in the present study. The statistical evaluation showed that the ANN model was more effective and precise than the other models in predicting the compressive strength, with 2.8% AA error, and strain at ultimate strength, with 6.58% AA error, of concrete-filled plastic tube tested under axial compression load. Similar lower values were obtained for the NRMSE index.


Main Subjects