Prediction of dynamic response of fluid in elevated water tanks using artificial neural network model

The objective of this study is to investigate the adequacy of Artificial Neural Networks (ANN) as a method to determine the dynamic response of fluid in elevated water tanks. For this purpose, an ANN models were proposed to estimate the hydrodynamic pressure in bottom of container and sloshing of water surface. ANN models were developed, trained and tested in a based MATLAB program. Nonlinear dynamic analysis using Finite Element Application (FEA) based ANSYS was used to generate training and testing set of ANN models. In the ANN models, a multilayer perceptron (MLP) with a back-propagation (BP) algorithm was employed using a scaled conjugate gradient. The data used in the ANN model are arranged in a format of three input parameters that cover the time history of earthquake horizontal acceleration, container ceiling displacement and base shear force.The performance of the new ANN model is compared with ANSYS results. The comparison indicates that the ANN model has strong potential to estimate hydrodynamic pressure. It was demonstrated that the neural network based approach is highly successful to estimate response of fluid subjected to earthquake without using complex fluid elements.