Title :
Quantitatively modeling multiple phase transformations in metals using generalized Hopfield nets
Author :
Schmitter, Ernst D.
Author_Institution :
Univ. of Appl. Sci., Osnabrueck, Germany
Abstract :
A cube of M=N3 neurons representing cells in a metal is used to model multiple phase transformations. After parameter fitting to experimental data the Hopfield net is able to predict industrially relevant processes quantitatively including their local behaviour
Keywords :
Hopfield neural nets; grain boundaries; metals; phase transformations; physics computing; solid-state phase transformations; generalized Hopfield neural nets; metals; multiple phase transformations; quantitative modeling; Aggregates; Entropy; Ferrites; Fitting; Neural networks; Neurons; Predictive models; Solid modeling; Steel; Temperature;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.830779
