IDENTIFICATION OF INDENTATION DIAGRAMS MODELS WITH MULTILAYERED NEURAL NETWORKS
Kruglov Igor Aleksandrovich
National research nuclear university “MEPhI”
PhD in computer science
AbstractThis article discusses a problem of neural networks based identification of indentation diagrams models obtained for four different steel types. Each experimental diagram is identified by three approximation coefficients and steel samples are characterized by two parameters of corresponding stress-strain curves (yield stress, rate of strain hardening). It is shown that neural networks with one hidden layer are able to provide a solution of the problem with sufficient generalization degree for available experimental data.
Category: 05.00.00 Technical sciences
Article reference:
Identification of indentation diagrams models with multilayered neural networks // Modern scientific researches and innovations. 2016. № 4 [Electronic journal]. URL: https://web.snauka.ru/en/issues/2016/04/66138
View this article in Russian
Sorry, this article is only available in Русский.
Artice view count:
Please waitAll articles of author «Круглов Игорь Александрович»