PREDICTION ACCURACY OF ENVELOPE CURVE FOR POST-INSTALLED ANCHORS BY MACHINE LEARNING WITH DECISION TREE AND NEURAL NETWORK

Abstract

Recently, artificial intelligence has been used in various fields, however researches of predicting load-displacement relationships for structural members are shortage. In this study, the shear force – shear displacement (Q -δS) relationships of post-installed anchors were predicted using machine learning with Decision Tree and Neural Network. As a result, the prediction results by Neural Network were the most accurate of the applied four methods. In addition, the prediction results of the Neural Network were compared with the evaluation results of the FEM analysis and Dowel model, which are the conventional methods. Finally, Neural Network was the most accurate algorism.