Development of Artificial Neural Networks to Simulate the Process of Dichloroethane Dehydration in the Statistica Software Program

Abstract

This paper discusses the application of artificial neural networks for simulation of technological processes. Stages of neural networks designing and training in the Statistica software package as exemplified in the source data of the technological process of dichloroethane dehydration were considered. After training the neural network this program allows to build histograms, scatter grams, graphical representation of the teaching selection on the three-dimensional surface, graphical representation of the training sample on the coordinate plane, etc. These charts help the user clearly see the process of neural network training and evaluate its effectiveness.