The paper presents a primary study of the latent space structure of neural networks trained for semantic segmentation. Segmentation was performed in a controlled environment of three classes of colored rectangular shapes. The classic autoencoder and U-net like architectures were chosen as reference architectures. To study the structure of the space, a combination of a perceptron that linearly separates classes and the compression algorithms UMAP and PCA was used. As a result, a tool was obtained for evaluating the quality of a neural network based on the degree of separability of classes in the latent space of the network.

The modes of operat ion of the steam turbine stage in non-steam and motor modes are considered. Analyt ical models of the operat ion of the stage with and without act ive power generat ion are obtained and analyzed calculat ion expressions are given for quant ifying the maximum allowable total power of the turbine stages when cooling steam flows into the flow part of the turbin e from the generator safety condit ion during its emergency disconnect ion from the network. Based on the thermal and balanceequat ions, acondit ion is obtained that excludes the possibility of external steam flows into the intermediate stages of the turbine affect ing the parameters of the steam flowing through the flow part

This survey covers the literature related to information communities in mutually complementary areas: the formation of information communities in social networks and some applied aspects of identifying and analyzing information communities in social networks. First, mathematical models describing the formation of information communities under uncertainty are considered. Among these models, the most relevant ones are the mathematical models of opinion/belief dynamics reflecting any changes in the beliefs of nodes under the influence of other network nodes and significant effects (in particular, the preservation of differences in beliefs and the divergence of beliefs) that lead to the formation of information communities. In part I of the survey, the concept of an information community is first presented. Then infor-mation processing and decision-making by an agent in a social network under external uncer-tainty are outlined. The factors influencing the formation of information communities in the network are highlighted, and the basic models of Bayesian agents and their extensions are investigated.