Integrating Artificial Neural Networks with Finite Element Method for Predicting Structural Responses of Plates with a Central Circular Hole Subjected to Uniform Tensile Traction

In this study, the structural responses of plates with a central circular hole under uniform tensile traction were predicted using a hybrid approach integrating artificial neural networks (ANNs) with the Finite Element Method (FEM). The FEM simulations were performed to generate a dataset, including various parameters such as hole radius, plate width, and applied load. The simulation outputs such as maximum displacement, maximum stress, and localized responses at specific points of interest. To combine the ANNs with the FEM, the simulation data obtained from FEM were used to be the training data for the ANNs. The multi-layer feedforward neural network was used, which was trained using the Levenberg-Marquardt method, to effectively capture the complex relationships between the input parameters and structural responses. The ANNs results were compared with the FEM simulation results to evaluate the accuracy of ANNs. The results show that the predicted data from ANNs model were consistent with the simulation results using FEM, indicating that the combination of ANNs and FEM simulation was effective in predicting the structural response accurately.