Математическое моделирование, компьютерный и натурный эксперимент в естественных науках
Электронный научный журнал

Технические науки
USE OF A NEURAL NETWORK FOR CONSTRUCTING AN ALGORITHM FOR STABILIZING AN UNMANNED AIRCRAFT OF A HELICOPTER TYPE
Sidorova M. A. 1, Barulina M. A. 2, Efremov M. V. 3

1. Institute of Precision Mechanics and Control of the Russian Academy of Sciences, Saratov, Russia, Research and Production Enterprise Antares, Saratov, Russia
2. Institute of Precision Mechanics and Control of the Russian Academy of Sciences, Saratov, Russia,
3. Research and Production Enterprise Antares, Saratov, Russia

Abstract:

At present, statistical approaches are used to predict the time of failure-free operation of radio-electronic equipment (REA) operated as part of spacecraft (SC). Such forecast methods have shown their effectiveness for mass-use products, but they can be insufficient for unique non-mass-use devices, such as those for the use in space. An alternative approach is a causal (physical and technological) approach to predict the probability of no-failure operation time. In contrast to statistical approaches, the causal approach is based on the analysis and assessment of precisely the causes and mechanisms of failures. But this approach does not completely exclude the random nature of failures either. In this article, an attempt to adapt the existing methods for predicting the reliability of electronic equipment for space applications was made considering the modern level of design and manufacture of the unique devices for space. ____________________________________________________________________________________________________________________________________________________ https://doi.org/10.24412/2541-9269-2021-1-17-24

Keywords: reliability, fiber optic gyroscope, Weibull distribution, statistical approach, physico-technological approach.

Сайт работает на RAE Editorial System