Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: International Conference on Aviamechanical Engineering and Transport (AVENT); Irkutsk Natl Res Tech Univ, Irkutsk, RUSSIA; Irkutsk Natl Res Tech Univ, Irkutsk, RUSSIA
Год издания: 2018
Ключевые слова: induction soldering, waveguide path, measurment error classification, flux, neural networks, joints
Аннотация: The article deals with the problem of classification of non-normative errors in measuring instruments. The use of such high-tech methods for creating integral joints, like induction soldering, electron beam welding, diffusion welding, is complicated by the presence of errors in measuring instruments associated with high temperatureПоказать полностьюs, the use of contactless temperature sensors, different thickness of the connected parts, different emissivity coefficients and human factor. Errors that arise when managing the process of creating permanent joints have a significant negative impact on the quality of products in various areas of engineering. In order to adequately control the technological process, it is necessary to identify and classify all types of errors in measuring instruments related to both the features of the technical means of measurement and the human factor. To solve the problem of classifying non-normative errors in measuring instruments, this article proposes the use of intelligent methods. The most suitable, effective and powerful means of solving the problem of non-normative errors classification is the use of artificial neural networks, which make it possible to develop the most effective control actions to compensate the arising non-normative errors.
Журнал: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE: AVIAMECHANICAL ENGINEERING AND TRANSPORT (AVENT 2018)
Выпуск журнала: Vol. 158
Номера страниц: 432-437
ISSN журнала: 23525401
Место издания: PARIS
Издатель: ATLANTIS PRESS