Размер:
AAA
Цвет: CCC
Изображения: Вкл.Выкл.
Обычная версия сайта

Search

Search:

Search by
Query string

Results:

Vol. 3 (2000 year), No. 1

Dranitsa Yu.P.
On one method of non-stationary dynamic systems and processes modelling

The correct method of an estimation of nonstationarity of the multidimensional dynamic processes and its taking into account when constructing linear mathematical models has been worked out. The theory of the method is based on approximation of the process by the linear differential equations system in direct and inverse time, and the knowledge of a covariance matrix is not required. The method is the generalization of the J.Burg theory, worked out for the one-dimensional and stationary case. The optimization of the algorithm has been executed by using properties of Teplitz matrixes and Levinson recursion.

(in Russian, стр.12, fig. 0, tables. 0, ref 9, MS Word 95, MS Word 95 35 Kb)

Vol. 3 (2000 year), No. 1

Dranitsa Yu.P.
On one method of incorrectly stated tasks solution

The method of the incorrectly stated tasks solution based on the Kalman theory filtration has been considered. The generalization of this method for the case of Markov processes of an arbitrary order in the correlated noise system has been made. The case, when for synthesis of the filter there is no necessary a priori information, has been considered, and the ways of the estimation of its parameters based on the statistical processing of the "on-location" information have been offered. The algorithm of parametrical adaptation of the filter to the processable data has been developed. The method is intended for the analysis and synthesis of complex technical and natural systems.

(in Russian, стр.7, fig. 0, tables. 0, ref 7, MS Word 95, MS Word 95 35 Kb)

Vol. 4 (2001 year), No. 1

Dranitsa Yu.P.
Modelling of one-dimensional dynamic processes with the purpose of preliminary processes

The theory of the preliminary processing of the one-dimensional sequences of data based on the linear dynamic model has been worked out. It has been shown that the problems of preliminary analyses and information transforming can be solved with the help of this method. The theory is intended for the analysis of one-dimensional time series. The possibilities of the method have been shown on a concrete experimental material. The designed method has a good theoretical substantiation and is based on the physical essence of the studied phenomenon.

(in Russian, стр.18, fig. 12, tables. 3, ref 8, MS Word 95, MS Word 95 238 Kb)

Vol. 5 (2002 year), No. 2

Dranitsa Yu.P.
Principles of neurowise modelling of geophysical objects and processes

In the work the task of the synthetic neurowise nets (SNN) application for modelling of geophysical objects and processes has been considered. It has been shown that this problem has a number of peculiarities caused by possible incorrect formulation of a task, noise in the observational data and shortage of experimental material. The methods of the SNN classic theory adaptation for geophysical tasks have been proposed. Some solutions have been shown for specific geophysical data.

(in Russian, стр.12, fig. 7, tables. 0, ref 12, MS Word 95, MS Word 95 161 Kb)

Vol. 12 (2009 year), No. 1

Dranitsa Yu.P., Dranitsa A.Yu.
Some aspects of interpretation basing on theory of line dynamic systems

The scientific-methodical foundations of multi-agent technologies and system dynamics methods integration have been developed. The generalized intelligent agent activity model in the virtual business environment of innovation evolution has been considered. The survey of existent agent architectures has been represented. The hybrid intelligent agent architecture with internal simulation subsystem constituting a simulation engine (a complex of system-dynamic models) has been designed. Implementation approaches to agent simulation engine based on system-dynamic models in the multi-agent environment have been proposed.

(in Russian, стр.7, fig. 0, tables. 0, ref 13, Adobe PDF, Adobe PDF 0 Kb)

Vol. 12 (2009 year), No. 2

Dranitsa Yu.P., Dranitsa A.Yu.
Some problems based on dynamic simulation

The paper contains the analyses of knowledge of the Murmansk region flora and approaches to protection of plant associations. The problem of rare plants' distribution in plant communities has been considered. Some types of plant communities to be protected have been proposed.

(in Russian, стр.10, fig. 2, tables. 0, ref 20, Adobe PDF, Adobe PDF 0 Kb)

Vol. 12 (2009 year), No. 2

Dranitsa Yu.P., Dranitsa A.Yu., Alekseevskaya O.V.
Spectral analysis based on the linear dynamic model

The paper considers some problems (tasks) in the framework of dynamic simulation including inverse and ill-posed tasks. It has been established that the classical solution of these tasks can be found in the context of so-called amplitude methods which resolution capability is in principle restricted by a length of a direct impulse wave. The reasons of weak resolution capability of amplitude methods have been analysed. Some proposals allowing to increase 1-2 orders of resolution capability of methods of some ill-posed tasks' solution have been made.

(in Russian, стр.8, fig. 0, tables. 0, ref 9, Adobe PDF, Adobe PDF 0 Kb)

Vol. 13 (2010 year), No. 4

Dranitsa Yu.P., Dranitsa A.Yu., Alekseevskaya O.V.
Setting and solution of primal problem of linear optimal filtering

In the work interpretation of time sequences from the viewpoint of dynamic modeling has been considered. The developed technique can be used for the decision of rather wide range of problems, namely: at designing linear optimum filters; separation (division) of mix of signals on components; research of dynamic parameters of time numbers; the frequency analysis and others. In particular, by working out linear optimum filters on a signal unique restriction – invariance in time of its statistician of the second order has been imposed. It can be noticed that the classical approach at the decision of this problem imposes on an investigated signal essential aprioristic restrictions which are actually carried out only for a very narrow class of data.

(in Russian, стр.7, fig. 0, tables. 0, ref 11, Adobe PDF, Adobe PDF 0 Kb)