Purpose of the work is design of methods and means for determination of transition’s probability of reconfigurable multiprocessor system (MS) aimed to control the complex objects. This transition can lead to dangerous state. Also as result of the work is enhance of system functioning results’ veracity. During exploitation due to the processors’ failures MS can went into dangerous state (that is state when is only one (any) of control functions can not be realized and without of those the control object went into dangerous state).

The aim of this research project was to study the asymptotic behavior of pseudo-regular functions and their analogues in stochastic analysis - generalized renewal processes, as well as applications of the results obtained to the theory of stochastic processes, statistics of stochastic processes, mathematical analysis and mathematical physics.
In this project:

Laws of energy induced diffusion phase-structural transformations in film functional-gradient materials with significantly different thermodynamic and crystal-chemical properties of the layers (V-Ag, Pd-Ho, Ni-Cu-Cr, etc.) were determined.
“Diffusion pump” effect was determined for investigated nanothickness materials: physical and chemical processes on outer surface thermodynamically determine diffusion phase formation in the bulk during energy impact in argon-, nitrogen-, oxygen-, hydrogen- containing atmospheres and in vacuum of 10-3 Pa, 10-7 Pa.

A complex study of heat transfer processes in constrained conditions in evaporative-condensation systems was carried out for the first time. A physical model of the boiling process on porous surfaces in conditions of limited space was developed and presented. Dependencies for calculating the heat transfer intensity under such conditions have been obtained. Such dependencies are necessary to create miniature cooling systems for microchips and powerful processors of electronic equipment.

Essentially new scientific knowledge has been obtained on the physico-materials science basics and the laws of the surface layers hardening of the light structural alloys by ultrasonic shock treatment (UST) at cryogenic temperatures. These regularities are due to the suppression of the dynamic return processes and dynamic recrystallization, as well as the effect of the dislocations annihilation under the action of the deep cooling factor.

A new model of information transformations was developed for remote thermal imaging from an aircraft or space vehicle. This model takes into account characteristics of background, targets, the optical system and the matrix detector, which allows us to explore ways to improve the technical characteristics of the system’s main blocks. The computational algorithms are based on spatial-frequency filtering in the signal propagation paths and allow performing the tasks of analysis and synthesis of systems with or without visual perception. The model is compatible with the current NATO standards.

For the first time physical representations and theoretical assumptions about the regularities of the influence of quasi-hydrostatic compression, as the most stringent conditions of deformation, on the composition, phase transformations, structure, strength and plastic properties of crystalline metal-like and quasicrystalline metal phases, including titanium hydride with a high hydrogen content and Al-Fe-Cr alloy with finely dispersed particles of quasicrystalline phases.

The research is aimed to creating of new methods for digital data processing in real time, transmission of the signals under the noise conditions, application of ontological decision-making systems for reducing the time of determining the critical human state and applying an emergency care.

The experimental sample of a new device for objective diagnosis of human hearing is developed - a broadband acoustic ear echo-spectrometer.

The theory of analysis and synthesis of adaptive electromechanical systems with vector-controlled electrical motors is developed and generalized. Proposed theory allows synthesizing automatic control systems with the properties of robustness to electrical motor parameters variations, thereby improving their dynamic performances and efficiency. Methods for robust adaptive estimation of immeasurable coordinates and identification of unknown parameters of electromechanical converters are developed.