Application of artificial neural network in the prediction of settlement of shallow foundations

Abstract

ABSTRACT

Besides assessing the bearing capacity, settlement is also an important parameter when designing shallow foundations. Therefore, this study presents the results of applying an artificial neural network model in the settlement of shallow foundations. An artificial neural network model was built and trained using the gradient descent optimization algorithm. A dataset including 284 results of shallow foundation load tests with different load levels, was used to train and test the model. The results of the study are compared with some research results of Meyerhof, D'Appolonia, showing that the artificial neural network allows for predicting the shallow foundation load more closely with the experimental results. The results of the study are a premise for the application of artificial neural networks in solving other problems in the field of construction.

Keywords: Shallow foundation settlement; SPT; artificial intelligence; artificial neural network.