ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 9. Vol. 25. 2019

DOI: 10.17587/it.25.572-576

S. E. Levin, First Deputy General Director, e-mail: levin@pkcc-ps.ru, Ya. N. Okrent, Chief Researcher, e-mail: namaste2003@gmail.com, Russian Corporation of Means of Communication — Software Systems, Moscow, 105005, Russian Federation
S. Ya. Nagibin, Ph. D., Professor, e-mail: nsy7@rambler.ru, N. E. Balakirev, Ph. D., Associate Professor, e-mail: balakirev1949@yandex.ru Moscow Aviation Institute (National Research University), Moscow, 121552, Russian Federation

Mathematical Model of the Technological Process of Styrene Production

The article presents a mathematical model of the functioning of the technological process of styrene production using neural network technologies. The use of a direct propagation neural network with one hidden layer trained on an experimental sample is considered. The algorithm for the formation of a neural network is proposed. The model is implemented as a software module. The results of forecasting the process of styrene production on real data and recommendations on the use of the developed model in the process of assessing the industrial safety of highly hazardous production processes are presented.
Keywords: industrial safety, artificial neural networks, mathematical model of styrene production, normalization of parameters, neural network training, method of back propagation of error, neural network formation algorithm, evaluation of prediction results

P.572-576

To the contents