Algorithmization of the failed subschemes localization process

Abstract

Approaches to the algorithmization of the process of localizing faulty subschemes of a wide class of electrical devices (in particular, electrical and electronic) are considered, appropriate algorithms are proposed, and the possibility of their application in practical applications is shown. In particular, it was noted that in modern conditions, with increasing requirements for the reliability of electrical devices and expanding the set of functions performed by them, not only the full feasibility of monitoring the operability of these devices at the production and operation stages should be considered, but also ensuring the potential possibility of diagnosing the developed devices (systems ) - at the design stage. At the same time, the observed complication of the methodological support of individual procedures and the whole diagnostic process as a whole is indicated, which, on the one hand, is connected with the need to carry out the most complete and comprehensive diagnosis, and on the other hand, determines a sharp increase in computing resources and labor costs for the implementation of a diagnostic experiment. One of the possible solutions to this contradiction is the formalization and subsequent algorithmization of diagnostic procedures, which ensures the automation of the latter and, as a result, reduces the time of diagnosis and improves its quality. The proposed algorithmic tools implement model-oriented methods for the localization of faulty subschemes of electrical devices (systems), in particular, the method of training and testing characteristics. A distinctive feature of these methods and the algorithmic support considered in the work is the use of models of the devices under study during a diagnostic experiment, which makes it possible to form and test functionally necessary (ideally, any) health hypotheses of the latter.