NEURAL NETWORK CLASSIFICATION OF STEEL TYPES FOR HARDNESS PROPERTIES ESTIMATION BASED ON KINETIC INDENTATION

Kruglov Igor Aleksandrovich
National research nuclear university “MEPhI”
PhD in computer science

Abstract
This paper focuses on neural networks based estimation of metals hardness properties using indentation data. An original type of a neural networks ensemble is employed which involves regularization. The study is based on experimental data of four steel types. For each steel type a separate instance of the ensemble is built. To extend the developed method to multiple types of metals it is proposed to use a neural networks classifier for indentation diagrams. It is demonstrated that for an observed sample of experimental indentation diagrams such classifier can be constructed from simple perceptrons.

Category: 05.00.00 Technical sciences

Article reference:
Neural network classification of steel types for hardness properties estimation based on kinetic indentation // Modern scientific researches and innovations. 2015. № 8. P. 1 [Electronic journal]. URL: https://web.snauka.ru/en/issues/2015/08/57116

View this article in Russian

Sorry, this article is only available in Русский.



Artice view count: Please wait

All articles of author «Круглов Игорь Александрович»


© If you have found a violation of copyrights please notify us immediately by e-mail or feedback form.

Contact author (comments/reviews)

Write comment

You must authorise to write a comment.

Если Вы еще не зарегистрированы на сайте, то Вам необходимо зарегистрироваться:
  • Register