The paper proposes a developed software application for automating the calculation of the technical characteristics of continuous ionites units with a fluidized ionite bed used in the water treatment system of the energy complex. The program is designed to calculate the structural and hydraulic parameters of an ion-exchange parallel-point filter used in water softening and demineralization schemes at water treatment facilities of heat power facilities. The necessary initial data, formulas are given, on the basis of which the algorithm is compiled, logical structures are developed and a software solution is obtained. The importance of work on ensuring high quality of water heat carriers for ensuring long-term and uninterrupted operation of the energy complex from the point of view of technogenic safety, as well as reducing capital and operating costs is shown. The possibility of using the developed software application in training specialists, as well as directly in working conditions, is described.

The article shows that the problem of fuzzy classification and the problem of separation of Gaussian mixtures have different theoretical solution models based on the results of the classical probability theory. A probabilistic model of the clustering and classification process based on the Bayesian approach is the joint distribution of multidimensional data points across clusters and cluster prototypes. The model is built for datasets with a normal distribution. Unknown membership values for each data point are assumed to be a random variable with some empirical distribution. The cluster prototype is also assumed to be a normally distributed random variable. The model is compared with the Gaussian mixture separation model and the main difference between them is highlighted: in the Bayesian fuzzy classification model, the product of all component probabilities is used, which means taking into account the possibility of the xn value falling into different clusters. The computational algorithm of this model may have a wider application compared to the traditional fuzzy c-means approach, due to the lack of restrictions on clustering parameters. Some numerical results are presented that confirm the model’s operability for an artificial data set both with parameters within the traditional limitations of the c-means method and for an extended range of parameter values.

No abstract

In this work, data on the development of such an important section of Electrical Engineering as “Electrical conductors and methods for their manufacture” are gathered together. The information collected will allow you to compare different materials suitable for the manufacture of electrically conductive structures. The paper also has a history of the development of this section, as well as a patent study of relevant and unusual methods for the manufacture of electrical conductors.