Розглянута можливість застосування інформаційної структурної технології визначення показника структурної відмовостійкості для підвищення ефективності планування технічного обслуговування і ремонтів технічних об’єктів. Технологія базується на побудові морфологічної моделі об’єкта та дозволяє отримувати значення величини ймовірності безвідмовної роботи системи при структурній подобі її моделі нейроподібній мережі Запропоновано механічним аналогом повторного навчання мережі вважати ремонт технічних об’єктів.
For systems with loaded backup, the design process involves a large number of discrete selection operations among various types of components based on their reliability or weight in the overall fault tolerance of the system. To calculate and design objects of chemical engineering, the calculation of reliability requires a sufficient amount of initial data. These data can be obtained from directories, based on the results of long and costly tests. Very often these data are extremely difficult to obtain or they simply are not available. However, the lack of such information or the inability to obtain it can not serve as a reason for refusal to carry out an assessment of the reliability of the object at any stage of its life cycle. One of the possible ways of obtaining information for determining the optimal method of the corresponding structural scheme is the use in the design and development of the design of chemical equipment information structural technology, which allows you to make a certain assessment of the probability of trouble-free operation of complex systems with loaded backup and predict the state of the system in the process of its degradation on the basis construction of information morphological model of complex object. The aim of the work is to increase the effectiveness of the design and operation of technical objects through the effective organization of repair works using the information morphological model. The technology is based on the construction of the morphological model of the object and allows obtaining the value of the probability value of the fail-safe operation of the system under the structural similarity of its model of the neural network. The mechanical analogue of re-training of the network is considered as the repair of technical objects.
