Scientific journal
Scientific Review. Technical science
ISSN 2500-0799
ПИ №ФС77-57440

CREATION OF NEURAL NETWORKS TO IDENTIFY THE THERMOPHYSICAL PROPERTIES OF WATER AND STEAM

Andreev V.V. 1 Orekhova E.E. 1 Abramov A.A. 1
1 Nizhny Novgorod State Universiry n.a. R.E. Alekseev
In this paper considered the application of artificial neural networks (hereinafter NN) to the solution of the problem of determining the thermophysical properties of water and steam at saturation line. Were created Ann for the definition of temperature, enthalpy, entropy and specific volume of water and steam at saturation line on known pressure. The comparison of the results of determining the thermophysical properties of water and steam using the networks of different types, estimated accuracy of determination of unknown parameters. Considered several different types of neural networks and identified the types of networks that are most suitable for the task solution. The study found that the network type RBF decides to solve the problem better than a chain-type multilayer perceptron. The subject of further research assumes determination of optimum parameters of a network such as the RBF for a solution of the problem.