engleski

Reducing computational time for FEM post- processing through the use of feedforward neural networks

With the recent surge in neural network usage, machine learning libraries have become more convenient to use and implement. In this paper the possibility of using neural networks in order to faster process displacements obtained from nite element calculation and replace existing post- processing procedures is investigated. The method is implemented on 2D membrane nite elements for their relative simplicity. A speed up is observed in comparison to traditional methods of post- processing. Possible further applications of this method are also presented in this paper.

Machine Learning, FEM, Postprocessing

nije evidentirano