The impulse fields reflected from layered medium with losses are analyzed by artificial neural network in time domain directly. The normal incidence of plane wave with Gaussian time form on layered medium is considered. The reflected electromagnetic field is obtained by FDTD method. Initial data for neural network are the values of amplitude of electrical component of reflected field chosen with constant time step. As an example the network is trained to determine the thickness of one of the layers of the medium.

The molecular dynamics simulation of the evolution of clusters of copper in a vacuum under low-energy argon ion bombardment has been performed. The clusters consisting of 26, 78, and 390 atoms have...

Molecular-dynamics simulation of the evolution of bipartite bimetallic clusters consisting of 390 atoms during bombardment by Ar_n (n = 1, 2, 13) clusters with initial energies from 1 eV to 1.4 keV is performed. Binary Cu–Au and Ni–Al clusters consisting of equal atomic fractions of corresponding elements were used as a target. As a result of simulation, the temperatures, changes in the potential energy, sputtering yields, and intensities of collision-stimulated displacement of atoms through the interface of monometallic parts of binary clusters, depending on the size and energy of incident particle, are obtained.

The molecular-dynamics simulation of free bipartite Cu–Bi clusters and, for comparison, Cu–Au clusters consisting of 390 atoms equally divided between the corresponding monometallic parts under bombardment by Ar, Ar₂, and Ar₁₃ particles with initial energies ranging from 1 to 1400 eV is carried out. Two sets of computer experiments for the 5- and 100-ps evolution of the system are conducted. The sputtering yields of single-component parts of the cluster and the number of polyatomic fragments of the sputtered material are obtained. The dependences of the potential energy and temperature on the energy and sizes of the bombarding particles are plotted. The evolution of the sputtering of bimetallic bipartite clusters from a collision cascade to the thermal one is shown; the thermal sputtering in this case is due to atom evaporation from cluster surfaces caused by high temperatures.

The subject of the research is the methods of constructing and training neural networks as a nonlinear modeling apparatus for solving the problem of predicting the energy consumption of metallurgical enterprises. The purpose of this work is to develop a model for forecasting the consumption of the power system of a metallurgical enterprise and its experimental testing on the data available for research of PJSC "Dneprospetsstal". The following tasks have been solved: analysis of the time series of power consumption; building a model with the help of which data on electricity consumption for a historical period is processed; building the most accurate forecast of the actual amount of electricity for the day ahead; assessment of the forecast quality. Methods used: time series analysis, neural network modeling, short-term forecasting of energy consumption in the metallurgical industry. The results obtained: to develop a model for predicting the energy consumption of a metallurgical enterprise based on artificial neural networks, the MATLAB complex with the Neural Network Toolbox was chosen. When conducting experiments, based on the available statistical data of a metallurgical enterprise, a selection of architectures and algorithms for learning neural networks was carried out. The best results were shown by the feedforward and backpropagation network, architecture with nonlinear autoregressive and learning algorithms: Levenberg-Marquard nonlinear optimization, Bayesian Regularization method and conjugate gradient method. Another approach, deep learning, is also considered, namely the neural network with long short-term memory LSTM and the adam learning algorithm. Such a deep neural network allows you to process large amounts of input information in a short time and build dependencies with uninformative input information. The LSTM network turned out to be the most effective among the considered neural networks, for which the indicator of the maximum prediction error had the minimum value. Conclusions: analysis of forecasting results using the developed models showed that the chosen approach with experimentally selected architectures and learning algorithms meets the necessary requirements for forecast accuracy when developing a forecasting model based on artificial neural networks. The use of models will allow automating high-precision operational hourly forecasting of energy consumption in market conditions. Keywords: energy consumption; forecasting; artificial neural network; time series.