Author:
Vinogradov M M,Zabirov A R
Abstract
Abstract
The article presents the construction of an artificial neural network to determine the temperature of destruction of a vapor film in subcooled and saturated liquids of water-ethanol mixtures. To train the neural network, the results obtained on cylindrical samples of stainless steel, copper and nickel are used. In total, about 260 experimental points were used, which is sufficient to build a specific computational model. This article discusses a model of a neural network of the multilayer perceptron type. The trained neural network model shows a greater generalizing ability than the theoretical model for determining the temperature of vapor film destruction.
Subject
General Physics and Astronomy