Bài báo khoa học

The aim of this study is to develop an efficient hybrid machine learning (ML) model, which combines the genetic algorithm (GA) and artificial neural network (ANN) for rapidly calculating the load-bearing capacity (LBC) reinforced concrete driven piles. An extensive database including 470 static tests is collected to train the hybrid ML model. The predicted results of the GA–ANN model in this study are compared to those of the pure ANN model. Statistical indicators containing the coefficient of determination (), root-mean-squared error (), and are determined to assess the prediction performance of the ML models. The comparison emphasizes that the GA–ANN model predicts the LBC of the pile accurately with a very high value of 0.99 and small of 49 kN. Furthermore, the effects of input variables on the predicted LBC are evaluated. Finally, to apply the ML model, a graphical user interface