The paper describes modern techniques of parametrical estimation of the failure rate leading function and the assessed failure rate under the assumption of incomplete recovery using Kijima models. Models are compared among themselves and with the models assuming full or minimal recovery of a component (homogeneous and nonhomogeneous Poisson processes respectively). For estimating model parameters, the method of maximum likelihood function is used. For estimating the failure rate leading function given in implicit form, the most popular method of modeling is the Monte-Carlo method. The paper also offers an example of application of the developed technique for calculation of reliability characteristics of components, which are part of the regular equipment of NPP power units.

The paper presents description of the developed measures for improving the quality of the system Atlant and reliability of its operation.

The paper offers analytical models, which allows obtaining expressions for determining the reliability indices of networks consisting of identical elements. For networks consisting of recoverable elements, the mean time between failures and the mean time to repair in steady-state operation are considered as reliability indices. In assessing the reliability of networks of non-repairable elements, such an index is the mean time to failure.

The paper is devoted to the problem of redundancy of generating capacities of the Unified Electric Power System (UES) of Russia. The paper has considered approaches to the solution of the problem used during the Soviet period, and now, peculiarities of redundancy in market conditions. General principles for redundancy used in Russia and abroad in a vertically integrated hierarchical structure of electric power management in a market environment are presented in this paper. Recommendations are given for completion of various individual norms regulating the procedure of selection and deployment of redundant capacities in the UES, and for the development of optimization models for solving problems of redundancy in a multiregional power systems association.

The paper studied the accuracy of transmission of vital messages for railway signalling and train controls systems using threshold fragmentation methods and antinoise coding. A scheme for the implementation of the transmission method of formalized messages with allocation of outer coding (generation of messages based on the principle of secret sharing) and internal coding (antinoise coding of fragments) is presented. The dependences of the probabilities of missing messages and receipt of a false report from the parameters of the fragmentation threshold scheme and antinoise coding with error detection and their correction are analyzed. It is shown that by changing the relation between the parameters of the external and internal encoding we can affect the relationship between the probability of missing a message and receiving a false report in accordance with the requirements made for the transmission of vital messages.

The given paper considers the mathematical model of dependability for a complex “facility of protection – safety system” with a periodically controlled safety system. A random fuzzy process of recovery is used for construction of a dependability mathematical model, which takes into account the uncertainty of model parameters. A fuzzy estimation has been obtained for the mathematical expectation of a complex’s failurefree operating life, and the calculation of mean time to failure of a complex “facility of protection – safety system” has been exemplified.

The paper presents a method for assessing risks associated with fractures and defects of side frames of freight car bogies. The construction of a risk matrix as per manufacturers of side frames, including the assessment of the impact and the frequency of events related to the fractures and detected defects of side frames, is considered. The proposed method is aimed at providing support for making management decisions as to further operation of freight cars on the basis of risk management.

A descriptive model and an expert-formal method have been developed for definition of a risk area – types of facilities most exposed to anthropogenic disturbances, as part of a complex object of protection, for the specified strategy of protection. The explanatory capacities of the model have been demonstrated through the analysis of protection strategies and consequences of their violations exemplified by a threat of terrorist attack.

BOOK BY I.B.SHUBINSKY STRUCTURAL RELIABILITY OF INFORMATION SYSTEMS.

BOOK BY I.B.SHUBINSKY FUNCTIONAL RELIABILITY OF INFORMATION SYSTEMS.