Problems of Artificial Intelligence №2016'1(2)

A. S. Minenko

AXIALLY SYMMETRIC FLOW

Summary: An axially symmetric flow with a free boundary within a sufficiently long domain is investigated. This problem is different from the ones considered earlier, that on part of the free boundary the nonlinear Bernoulli’s law is set in kind of inequality. The problem of the existence of boundary value problem is equivalent to minimum problem of some functional with a variable domain of integration. Analyticity and monotony of free boundary are proved. The approximate solution of problem, built by Ritz method, converges to exact solution in .

Keywords: differential equation, free boundary, functionality, optimization, numerical methods.

V. N. Pavlysh, G. I. Turchanin, O. A. Tikhonova

MODIFICATION OF COMPUTER METHODS OF PRESENTATION AND ANALYSIS OF GEOTECHNICAL INFORMATION

Summary: The article considers the problem of improvement of presentation ways and mathematical analysis methods of geotechnical information about mine rock massif state for computer support of decision on mine development systems and functional providing of mine workings.

Keywords: information, method, analysis, geotechnology, decision, algorithm.

A. V. Sosnitsky

INTELLECTUAL SCIENTOMETRICS

Summary: The paper is dedicated to the problems of modern scientometrics that is based on the scientific work information properties, which do not allow adequately assessing fundamentally intellectual properties of knowledge. We offer to solve it by the universalization of scientometrics and its transfer into the system of intellectual indicators, in which concepts are the units of knowledge measurement. Since the metaconcept of intelligence is not defined in modern science, the so-called universal theory is applied to this purpose, which for the first time deduces it from the highest properties of our universe. For the first time the universal metaclassification of knowledge is deduced in accordance with the requirements of the Supreme Certifying Commission. The methods of intellectual scientometrics implementation in science, engineering and education are offered.

Keywords: information scientometrics, intellectual scientometrics, universal theory, universal model, the system of universal concepts, the universalization of knowledge.

S. S. Antsyferov, A. S. Sigov

TECHNOLOGICAL BASES OF CONSTRUCTION INTELLIGENT SYSTEMS

Summary: The article defines the basic principles of the practical implementation of intelligent systems with the help of computer technology. Special attention is given to the technology of nano-computer development. It was noted that the adoption of the nanometer range requires the development of new physical principles and technologies for production of component basis. The unity of bioinformatics and nanobiotechnology has great prospects allowing to create intelligent implantable nanosystems for monitoring the state of the body at the cellular level.

Keywords: supercomputer, silicon technology, volatile memory, reconfigurable system, ultradense information recording, nonsilicon technology, carbon nanotubes, component density, quantum computer, qubit, biocomputer, massive parallel circuit, memory capacity, processing speed.

V. А. Коsuk, I. V. Nikitina

INNOVATIVE PROCESSES MODELING EQUIPMENT OPTIMIZATION OF COAL INDUSTRY ENTERPRISES

Summary: The article considers the innovative approach to the problem of improvement of coal-mining industry equipment in the software product the All Fusions Win, which results in increasing of labour productivity and competitiveness.

Keywords: innovative approach, target flow, decomposition, notation IDEF0, All Fusions Win software product.

V. N. Kotenko, I. N. Polishchuk

METHOD TO INCREASE EDGE DETECTION QUALITY USING PARTICLE SWARM OPTIMIZATION ALGORITHM

Summary: New algorithm was designed on the specific objective to improve edge detection performance on images. Existing edge detection algorithms performance was researched on practice. New approach of edge detection using particle swarm optimization was proposed to increase localization accuracy on heavy noised images.

Keywords: image processing, edge detection, homogeneity, Kenny algorithm, Particle swarm optimization, performance.

V. J. Przekop

SYMMETRIC DISCRETE ORTHOGONAL FUNCTIONS OF FINITE-DIFFERENCE SERIES EXPANSION

Summary: The article considers a new group of symmetric discrete orthogonal functions allowing us to represent time series (discrete function) in the form of the sum of finite differences. These functions may be applied in the digital signal processing, data-compression system, discrete signal analysis, calculation of trends for time series.

Keywords: discrete orthogonal functions, orthogonal discrete transformations, finite differences, digital signal processing.

V. Ju. Shelepov, A.V. Nicenko

SEGMENTATION OF SPEECH SIGNAL ON THE ASSUMPTION ABOUT ITS PHONETICAL STRUCTURE

Summary: The article contains algorithms for segmentation of speech signal corresponding to the beforehand known word or continuous speech phrase. One part of the algorithms corrects and qualifies the a priori segmentation. The other algorithm, using the so-called DTW equalization-path, is applied directly to the signal. In this case the source of marks is the etalon of word or phrase which is synthesized from the diphone etalons. In combination with a priori recognition we obtain unconditional a posteriori segmentation.

Keywords: a priori segmentation, conditional segmentation, wide phonetic classification, mark, equalization-path, diphone-base modification.

Print