USING NEURAL NETWORKS IN THE PROCESS OF OBTAINING SPONGE TITANIUM

Abstract

This article is devoted to the planned development of a software module based on deep learning of neural networks in the workshops of the JSC "Ust-Kamenogorsk Titanium and Magnesium Plant" (JSC "UKTMP"). Now, the production of sponge titanium in UKTMP JSC is accompanied by a manual mode of work of specialists, which does not exclude human error of specialists in production. One of the ways out of this situation is the developing of full automation of the process of obtaining sponge titanium based on deep learning of neural networks, as well as the implementation of a MES system. The article indicates the method of deploying neural networks for analyzing the accumulated data of the technological process for obtaining sponge titanium.