Vol 25, No 6 (2019)

When using the maximum principle to find the optimal control, it is necessary to solve a two-point boundary value problem for a system of ordinary differential equations, which is more complex than the initial one. However, in some important cases, for example, when optimizing linear systems with a quadratic functional, the two-point problem can be transformed to the Cauchy problem. Initially, the traditional problem is solved with the transfer of boundary conditions, and then after phasing of the parameters of the problem, the corresponding fuzzy problem is solve. When transferring conditions for a two-point boundary value problem from the left on the right end of the trajectory, the solution consists of the following steps. An initial task is formed on finding vectors of conjugate variables. An initial problem is defined for determining a transferable vector. A system of algebraic equations is solved for the state vector components missing at the right end. The transition to a fuzzy boundary-value problem is performed by fuzzing the elements (parameters) of the matrix for the classical problem and applying the expansion principle to it when its Seikkala and/or Buckley-Feuring solution is obtained using the transition matrix. The example is given.

In recent years the increasing attention of scientists and designers of technical systems has been acquiring the issues of improving methods for assessing the reliability and safety of these systems, in connection with tasks of increasing the values of these characteristics. The purpose of the article is to develop an automata-logical model of reliability of complex technical systems and corresponding logical methods for evaluating the reliability of such systems, which, unlike known ones, use not the traditional probabilistic reliability indicators, but deterministic logical indicators. In order to achieve this goal, the article suggests using the observed moments of successive failures and recovery of the elements of the technical system as initial data, and as the reliability characteristics of the system itself the moments of successive failures and recovery of this system In this case, the problem of estimating the reliability of a system is reduced to constructing its mathematical model in the form of automata logical functions expressing the moments of its successive failures and reconstructions through analogous moments of all its elements. This article is the second part of the work in which an automata-logical model designed to calculate the logical function of reliability of complex technical systems is developed in detail. The novelty of the work is the construction of an adequate logical model of the reliability of a complex system, which makes it possible to reduce the estimation of reliability of a complex technical system to the calculation of its logical reliability functions. In the process of calculation, the mathematical apparatus of logical determinants is used for the first time, which allows us to solve the complexity problem. In the article the logical model of reliability and methods of its investigation are developed in detail, allowing to introduce new indicators of reliability of complex technical systems that do not require for their evaluation the use of probabilistic methods and initial statistical data on element failures. On the basis of the developed logical model of reliability and methods of its investigation, the problem of constructing an automata system for reliability of systems is solved, which will allow to fulfill practical calculations of complex technical systems by methods of the theory of dynamic automata using the apparatus of logical determinants.

The ways to increase the productivity of computer-aided design based on the transition to a compact form of storage and processing of sparse matrices. The description of the most effective method of compact storage and processing of sparse matrices on the basis of a fixed row-column format is given. It is shown that a distinctive feature of the fixed format methods is the impossibility of including additional non-zero elements in the description, which excludes the possibility of direct application of these methods for information processing in solving systems of equations by any numerical method, due to the inevitable appearance of new non-zero elements in the process of this solution. To be able to take into account new non-zero elements in the compact description of the simulated system, it is proposed to use the index-numerical technology for the formation of such a description and a description of the implementation of data compression based on the division of the General procedure into two independent parts of the index and numerical analysis. The technique of construction of the software of systems of automation of circuit design on the basis of index-numerical technology of data compression is given.

The indicators of the quality of computational facilities that control a distributed multi-channel queuing system are given. The main indicator of quality, ensuring the use of computational tools in control systems, is the actual performance. It is determined by control or functional tasks. For the proposed projects of computing means or projects of their modernization as a result of tests or by an expert, the expected quality characteristics are found. According to them is the rating of such projects when they are used in a distributed multi-channel queuing system. Rating is found by associative comparison with known, proposed or conditional projects of a known rating. The rating may take the following values: high, high enough, taking into account possible improvements, medium successful, suggesting further development of scientific and technical research, medium, indicating a wrong or technologically inaccessible way to choose a solution, but deserving attention for lack of other choice, low at which development must be rejected.

The decision system is built on a logical neural network to assess the quality of the processor (microprocessor), a logical neural network to assess the quality of integration of the VS and VC, a logical neural network to assess the quality of the network interaction of computing facilities, and a logical neural network for expert and difficult to verify evaluations of the quality of computing means.

The task of effective selection of ground-based objects based on the results of the use of laser means of scanning them can be solved provided that the object is reliably selected against the background of a complex masking relief. The recognition result will be identification (assignment of the proper and unambiguous name to the object in question) or classification (determination of the object’s belonging to a given class with an estimate of their coordinates). Identification or classification of objects is made on the basis of an a priori dictionary of attributes and the alphabet of their division into classes. In the self-learning recognition system, the feature dictionary can be filled (corrected). The development of a generalized algorithm for the formation of a feature space required for reliable separation of image classes is one of the most important and timeconsuming tasks of creating a system with recognition, since in most cases the distinctive features do not have clear boundaries and cannot always be formalized. Selection of features can be carried out in the measurement space and the transformed space. Reducing the number of features reduces the cost of carrying out measurements and calculations, but can lead to a drop in the reliability of recognition. If the time for training and decision making is severely limited, then increasing the dimensionality of the attribute space may be the only means of increasing the reliability to the required level. Thus, from a practical point of view, the requirements of the minimum of the overall dimension of the problem of recognition and the maximum of confidence are in conflict. In this article, the author examines the task of forming an a priori alphabet of classes, based on an a priori compiled dictionary of object recognition features from the results of an airborne laser radar scan.

In order to modernize the method of ordering book publications, which is based on their ranking by the number of references to sources previously used by the authors of the university, it is proposed to assign a weight to each university link calculated by the author’s citation indexes of the referencing work that allows to define more precisely and to satisfy more information needs of the university.