комп’ютерне моделювання

This article discusses the practical aspects of applying logistic regression for binary data classification. Logistic regression determines the probability of an object belonging to one of two classes. This probability is calculated with the help of a sigmoid function, the argument of which is a linear convolution of the feature vector of the object with the weighting coefficients obtained during the minimization of the logarithmic loss function. Predicted class labels are determined by comparing the calculated probability with a given threshold value.

The effectiveness of the computer modeling use of wastewater treatment processes during the implementation of planned measures was proved on the example of reconstruction of wastewater treatment plant (WWTP) of one of the large cities of Ukraine. Modeling of technological processes of wastewater treatment was performed for six modes of operation of WWTP: according to the operating technological scheme; during the implementation of four stages of its reconstruction and after the completion of all works of the reconstruction of WWTP according to new technological scheme.

The widespread use of mathematical applications, which include differential equations solvers, and the increase in the speed of computing devices have led to a decrease in interest in operator methods, in particular, the z-transform. Nevertheless, the use of the z-transform capabilities allows the implementation of efficient high-speed computing schemes with high numerical stability. The need for this may arise in the case of real-time simulation or the synthesis of digital control systems.

The influence of chemical nature of modifier functional groups on the level of intermolecular interactions in the system "epoxy oligomer ‒ modifier", as well as the structure formation and dynamic viscosity of epoxy oligomer has been studied in detail. Modifying additives in low concentrations contribute to an increase in the degree of structure formation of the epoxy system by increasing the intermolecular interaction between the associates of the epoxydiane oligomer.

Structural models of system stabilizers of power grids that are used to improve damping of power system oscillations by controlling the excitation of synchronous power plants turbogenerators are considered in the article. Mathematical and structural models of such a system stabilizer for various orders of its transfer function according to the IEEE recommendations are proposed for implementation in computer design systems, in particular, for the computer analysis system of the DAKAR power grids.

This article surveys an analysis of dynamic processes of hoist of the excavator-dragline type ЕШ-15/90 using computer simulation, considering the effect of sagging of the rope. A comparison of the accuracy of the obtained computer models with and without the nonlinearity of its elasticity coefficient is considered. Structural models with the use the libraries of MATLAB + Simulink mathematical program are proposed to solve this problem.

On the base of computer modeling with calculation of the rock radioactivity was analyzed the behaviour of the heat fields of the central part of DDH in the beginning and in the end of the rifting stage. It was showed, that temperature conditions necessary for activity the process of eclogite metamorphism in the lower layers of the crust took place during all this stage. Another hand, for activizations of process of progressive stage of granulate metamorphism in the lower granite layers it was necessary the process of long heating of the crust during about 10 ml. years.

A method for estimating and predicting abnormal layer and pore pressure during drilling of oil and gas wells using finite element method is elaborated. Utilizing simulated stress-strain state of a real model of a massif it is shown the qualitative agreement of results with theoretical calculations. We identified evaluation and prediction of pore and layer pressures and reservoir in the rock massif.

Описана математична модель KWTA-нейронної схеми (“K-winners-take-all”), призначеної для ідентифікації К найбільших серед N змінних у часі дискретизованих сигналів, де <≤ NK1 . Встановлюється, що для забезпечення коректного функціо- нування моделі динамічний зсув вхідних сигналів протягом перехідних процесів повинен змінюватись набагато швидше ніж вхідні сигнали. Наведено відповідні результати комп’ютерного моделювання.