Пожалуйста, используйте этот идентификатор, чтобы цитировать или ссылаться на этот ресурс: http://mx.ogasa.org.ua/handle/123456789/9442
Название: STRUCTURED SYSTEMS OF FACTORS AND EXPERIMENTAL- STATISTICAL MODELS IN STUDIES OF BUILDING COMPOSITES
Авторы: Lyashenko, T.
Ключевые слова: mixture
factor domain
simplex
design of experiment
reduced polynomial
composition-process field
Дата публикации: 2021
Издательство: МММ, Одеса, ОДАБА
Серия/номер: том 3. вип. 1;с.47-61
Краткий осмотр (реферат): The introduction to the article presents the objects of research, for which the proposed approach to modelling and the types of the models are intended. These are high quality composite building materials, the multicomponent dispersed systems, with the components that could be multicomponent themselves. The need to use for their design the mathematical models of the dependences of structure parameters and properties of the material on the factors of composition and processes of production and operation is noted. Then the background of experimental-statistical (ES) models application in research and development of composite materials is considered. What they are needed for is emphasized. The peculiarities of these models are noted in contrast to the models of other classes, including the usual regression ones. The necessity of experiment design to build ES-models is also emphasized. Special types of ES-models are named, for the cases when it is necessary to consider linearly related factors ("mixtures"). The expediency of structurizing the system of all the factors under consideration and the corresponding ES-models is indicated to. The concept of a priori and a posteriori structuring of factor systems when modeling is proposed. Systems that could include subsystems of linearly related factors and subsystems of mutually independent factors separated out at the stage of a priori structuring have been called “mixtures, technologies" systems and classified by the type of factor domains. System designations are given. Special polynomial models developed for them are presented. For the analysis of modelling results, a posteriori allocation of the factor regions and the use of tools of composition-process fields methodology are proposed. The local fields of material properties in coordinates of one or another group of the factors and their transformation under the influence of the factors of another group can be analysed, using numerical generalizing indices of the local fields and the secondary models for these indices. The path from obtaining data of designed natural experiment for building the primary ES-models to the secondary models, for generalizing indices, based on the results of computational experiment, is schematically shown.
URI (Унифицированный идентификатор ресурса): http://mx.ogasa.org.ua/handle/123456789/9442
Располагается в коллекциях:МММ, том 3 Випуск 1

Файлы этого ресурса:
Файл Описание РазмерФормат 
STRUCTURED SYSTEMS OF FACTORS.pdf419,46 kBAdobe PDFПросмотреть/Открыть


Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.