Search

This paper presents the obfuscation code technique based on the two machine learning models. The models are based on the recurrently trained neural networks, represented as tensors and integrated into the application binaries.

This paper considers the mechanisms of the applying the subtasks prioritization on workflow performing by a distributed wireless computing system (DWCS) with a resources control mechanism based on the network-centric approach. The ordering by priority of subtasks allows to reveal the critical path and to speed up the execution of the subtasks. In this research, the setting of subtasks priority was performed according to the algorithm Heterogeneous Earliest-Finish-Time (HEFT). The experimental researches of the proposed approach are also presented in this paper.

The kinetics of electrode reactions that progress on silver in acid thiourea-citrate solutions has been studied. It has been established that the electrochemical surface polishing effect is reached in these solutions. This effect is conditioned by the formation of the salt oxide film on the anode surface. Silver is ionized through the passive film that contributes to the dissolution of microelevations and the surface microprofile is smoothed. The silver dissolution (polishing) rate is limited by the inhibition of electrochemical reaction at the diffusion stage. The silver polishing process is affected by the following factors, in particular, the solution composition and the concentration of its components, the surfactant (ethylene glycol), the solution temperature, pH, the working current density and the anode potential.

Semiconductor nanowires (NW) have received much attention for their unique properties providing exceptional flexibility in device engineering. In particular, gallium phosphide NWs demonstrate effective wave-guiding properties and broadband nonlinear frequency conversion, and find applications in green to amber optical range optoelectronics and photocatalytic devices. They also exhibit one of the broadest transparency range, high values of the nonlinear refractive index and good thermal conductivity that make them perspective for nanophotonics.

A synchronous generator has a significant impact on the stability and reliability of consumers' power supply. At present, the problems of emergency and operational control of power systems are being solved on the basis of computational models. The parameters of such models are determined using reference data or test data. The parameters of generator models are highly dependent on external factors. The presence of this dependence leads to a significant decrease in the accuracy of solving the problems of power system control. Currently, identification based on traditional telemetry systems or synchrophasor measurements is used to improve the accuracy of the power systems computational models parameters. This research aims to analyze the studies, the authors of which have developed methods for determining the parameters of a synchronous generator based on measurement data. In the research, both Russian and foreign works are analyzed and grouped. For each group, the advantages, disadvantages, as well as the area of application were identified. It was concluded that the existing methods for determining the parameters of synchronous generators from the measurement data are not able to change in accordance with the set of initial data. In addition, the research has shown that these methods require significant computing power. The adaptive model of a synchronous machine is proposed as a way to overcome these disadvantages.

This paper discusses the modeling of the process of interaction in traveling-wave tubes (TWTs) based on “coaxial-radial line”-type (CRL-type) slow-wave structures (SWSs). One of the most important tasks while designing TWTs is to achieve high output power over wide range of frequencies. The aim of this paper is to try to solve this problem by combining the broad passband which is exhibited by spiral SWSs and good heat dissipation of “chain of coupled cavities” -type SWSs. As a result of this work, an algorithm for calculating the characteristics of transparent multi-beam TWT based on the SWS of the CRL-type and a method for designing TWT data have been developed. The calculation of transparent TWT was carried out. The calculation results show that in such TWT the problem of obtaining high power in combination with good heat dissipation can be solved. In addition, the developed program allows designing both transparent and sectioned TWTs.

This paper is devoted to the analysis of the social risk during the operation and information processing in the complex potentially hazardous objects. There are considered the main factors, that have affect on social risk. In order to perform the risk assesment there was designed and learned the several neural network structures. The method of choosing of the neural network structure which allow obtain the smallest relative error of the result has been proposed and the most acceptable stucture of neural network was chosen. The results of risk assessment shows that the neural network are very close to the results obtained by the dedicated system for risk analysis, so the possibility to use the neural network for the risk analysis in this case is approved.

Two methods of disturbance-based energy analysis of synchronous generator rotor acceleration were considered. It was done for the purpose of stability analysis using the equal area criterion in the time domain. The current measurement was ensured by using the suggested method of correcting the secondary current of current transformer with saturated core. The methods performance was tested on the 4-bus model using the Matlab Simulink software. The results have shown that error values did not exceed 3% for saturated current transformers.

The article presents the development of a broadband quadrature signal generator (QSG) from 10 MHz to 6 GHz. When designing the QSG in broadband of frequencies, there are some difficulties when using SiGe BiCMOS technology. The problems of linearity and broadband matching are most challenging to solve when using traditional solutions: using a polyphase filter or a digital D-trigger. The article presents a structure that allows you to develop a broadband quadrature signal generator with a range of operating frequency from 10 MHz to 6 GHz, using the method of dividing the operating frequency band. At frequencies from 1 to 6 GHz, the quadrature signal is generated using a polyphase filter, and at frequencies less than 1 GHz using a frequency divider. It is shown that using this method, and the following characteristics QSG are achieved: the difference in the amplitudes of quadrature signals is less than 0.3 dB; the SWR is less than 1.6 in the entire frequency range; the transmission coefficient is not less than -2 dB with a 0 dBm heterodyne power; at a frequency of 6 GHz, P1dB is not less than three dBm. The experimental results are in good agreement with the simulation data.

Quadrature modulators and demodulators are widely used to create modern wireless communication systems. It is crucial to ensure the high quality of the transmitted signals for the exchange of information takes place without loss or failure. From the point of view of the spectral decomposition of the signal (Fourier series decomposition), the useful component of the spectrum must be much larger than other components. The most critical undesirable quadrature modulator output signal spectral components are a carrier (LO frequency) and spurious sideband. In the course of the research, using computer simulation, methods of their suppression were identified, based on minimizing phase, amplitude, and current imbalances in various nodes of the quadrature modulator circuit. The effectiveness of methods is confirmed by the achievement of suppression of parasitic components in prototypes of 50 dB or more.

A mathematical description of the influence of a true type logarithmic amplifier on the level of intermodulation distortion of a modulator/demodulator using a Taylor series expansion is presented in the article. Method of improving the dynamic parameters of microcircuits is described, and the results of spectral analysis of the modulator/demodulator using a logarithmic amplifier are presented.

The adaptive model of the synchronous machine for electric mode control with parameter selection based on operating mode data is presented in the paper. The structure of the model changes depending on the set of measurements. The mathematical description of the model core is given. The example of calculation for measurement results of electric mode parameters of the hydro-generator with full rated power 370 MVA during the real electromechanical transient process is conducted and explained. The inertia moment of the turbine’s and hydro-generator’s rotating mass was determined based on the calculation. The suggested adaptive model can be applied to emergency control tasks, including post-event emergency control and estimation of technical conditions of synchronous machines. It is shown, that response of adaptive model of the synchronous machine depends on measurement sampling rate. The adaptive model of the synchronous machine can be used for represent a complex load and an external equivalent of power system.

The accelerated synchrophasor measurements algorithm is presented. The algorithm allows to evaluate synchrophasors with less than one basic frequency cycle time delay. The reference points method was suggested to determine the synchronous frequency, with magnitude and phase being calculated by means of extrema interpolation. Tests were carried out using the simulation data and the real data of the 110 kV, 220 kV, and 500 kV grids steady states and transients. Requirements for time delay of the suggested algorithm and sampling rate of the initial measurements data were formulated. The total vector error minimum was selected as the optimality criterion. Since the algorithm was tested on the real data without actual reference values, the alternative method of defining the total vector error was proposed. The suggested algorithm can be applied to protection & control systems based on the reactive action (the so-called “1-After” concept, when the set values of protection & control systems are calculated after a disturbance), development of dynamic power system models and generators technical state assessment.

Nowadays, the algorithms of the windowed digital signal processing are widely used for the development and introduction of the phasor measurement units. The specific feature of these algorithms is the presence of spikes at the beginning of the transient process due to the inclusion of both steady-state and transient processes into a window span. As a result, the error of the electric mode parameters acquisition during the data accumulation period rises. This effect can be mitigated by shifting the window to points after the beginning of the transient process. In this paper, we present an effective algorithm for real-time determination of the inception point of the transient process. The algorithm was tested on model-generated and real signal samples. The criterion is suggested for precision enhancement of the algorithm by utilizing beginning points of transient processes of phased and linear voltages and phased currents. The error of prediction for the real samples testing does not exceed 0.001 seconds.

In recent years, the technology of synchrophasor measurements has been actively introduced in power systems around the world. When managing the power system and predicting its modes, the important tasks are to check the reliability of data and estimate the state of the power system. The introduction of synchrophasor measurements, on the one hand, makes it possible to solve the problem of state estimation more efficiently in comparison with traditional telemetry. On the other hand, previously used data validation technologies are not suitable for their use with synchrophasor measurements due to the new properties of these measurements. It should also be borne in mind that these two tasks are interrelated. This study provides an overview of recently proposed power system state estimation and data validation methods based on synchrophasor measurements.

The technology of thermal diffusion bonding of Yb:YAG/Sa composite disk active elements (AE) has been developed. The first experiments demonstrate a good optical quality of composite AE, suppressing of ASE and imrooving of thermal effects.

We demonstrate regenerative amplification of chirped pulses from sub-nJ to multi-mJ level with CW pumped Yb:YAG thin-disk laser head. Multipass amplifier with quasi-CW pumping is developed to provide further amplification to 200 mJ range and ~70 mJ is demonstrated in first experiments at reduced repetition rate.

Surface Activated Direct Bonding method is realized for manufacturing composites from dissimilar materials for high power lasers. Yb:YAG/Sapphire active element for thin-disk laser and TGG/Sapphire for high-power Faraday isolator is made for the first time. The advantages of new composites are demonstrated. Continuous-wave laser based on Yb:YAG/Sapphire active element with 49% slope efficiency and 320W average power is developed.

We present a comparative study of two amplifier modules for femtosecond laser system with up to 80W of average power at repetition rate 3MHz based on fiber seed laser and double stage single crystal fiber or thin-rod amplifiers. By using tapered geometry of a single crystal fiber developed in IAP RAS in second stage we obtained higher gain and efficiency compared with traditional cylindrical thin-rod shape single crystal fiber.