Engineering Mechanics

Proceedings Vol. 24 (2018)

ENGINEERING MECHANICS 2018

Copyright © 2018 Institute of Theoretical and Applied Mechanics of the Cech Academy of Sciences, Prague

 

Wróbel J., Kulawik A.
pages 941 - 944, full text

The article is a continuation of the considerations on the heat treatment of steel bars. In the paper, a method of automatic velocity choice of a heat source with the use of an artificial neural network is presented. The heat treatment process was performed for the long axisymmetric elements with variable diameters made from medium carbon steel. The two cases of change, linear and step, of the radius of the steel element are considered. It is assumed that the velocity of the heat source is modified in the process and is updated in real time according to the current diameter. The measurement of the diameter is performed at a constant distance from the heat source. The main task of the model is to control the assumed values of temperature at constant parameters of the heat source such as radius and power. The aim of the system, assuming that the heat source has constant values of power and radius, is to provide the established values of temperature on the surface of a quenching element. Therefore, the parameter of the process controlled by the artificial neural network is the velocity of the heat source. The artificial neural network is trained on the basis of results obtained from copyrighted numerical model of thermal phenomena (stationary task). To solve the steady heat transfer equation (axisymmetric task) with a convective term the finite element method is used. To verify the trained artificial neural network numerical simulations are performed based on the solution of the unsteady heat transfer equation.

Text and facts may be copied and used freely, but credit should be given to these Proceedings.

All papers were reviewed by members of the scientific committee.