Neural Network Modeling of Heat-Exchange Properties for Surface Intensification of Heat Exchangers

Abstract

The results and methods of neural network modeling of the average heat transfer during the intensification of a heat exchange surface are presented. The results of neural network modeling are presented for the following types of surface intensifiers: ring, spherical, cylindrical, oval-ditch recesses, and protrusions. The training data of the artificial neural network was formed on the basis of experimental data. For each type of surface intensifiers, graphs of the network test results spread relative to the actual values of the experimental matrix are given.