CORRECT STATISTICAL MODELING IN CONDITIONS OF INCOMPLETE INITIAL INFORMATION
Keywords:Statistical modeling, ill-posed problems, stable structure of statistical model, extended conception of orthogonality.
Context. Urgent problem of statistical modeling of the complex systems and processes in conditions of incomplete initial information has been considered.
Objective. The work objective is the use of the method of formalized obtaining of the structure of multifactor statistical model and stable estimation of its coefficient for obtaining highly precise statistical models of elastic deformations of technological system of a lathe.
Methods. When solving applied problems, the analysis of initial data on obtaining statistical models has shown that they are often constructed in conditions of incomplete initial information and the problem which is solved is incorrectly formulated one. In such conditions the model structure obtaining and its stability prove to be the problems in models construction. The author proposes an extended conception of orthogonality of the obtained model: the experiment design, model structure and structure elements of the model are orthogonal. The orthogonal structure of the multifactor statistical model allows obtaining statistically independent estimates of coefficients of the modeled function. Such a structure may be defined unambiguously with statistically significant coefficients. Normalization of orthogonal effects permits obtaining a maximally stable model structure, and, consequently, its coefficients. The problem will be well-posed.
Results. Application of the considered method of formalized obtaining of the structure of multifactor statistical model and stable estimation of its coefficients is used for obtaining accurate statistical models of elastic deformations of a steel work-piece processed by a lathe. A complete factor experiment has been fulfilled; the factors were as follows: cutting force, work-piece length and diameter, and a response – the value of elastic deformations of the system. Statistical regression moments ˆy1 and ˆy2 were constructed as the experiment result. In the structure of models the factors are presented by orthogonal contrasts. Statistically significant effects are introduced in the model structure under its formation. The checks of the obtained models by quality criteria have shown their high informativeness, stability, adequacy, statistical efficiency. Using the models on lathes with numerical programmed control allows decreasing the number runs of the cutting tool and, consequently, the time of work-piece processing.Conclusion. The results of the use of the extended conception of orthogonality and structure of the model of a complete factor experiment, when obtaining the models of elastic deformations of technological system of a lathe, have confirmed the great prospects of application of the considered approach, its effectiveness and expediency in constructing regression statistical models of complex systems and processes.
Voznesenskij V. A., Vy’rovoj V. N., Kersh V. Ya. i dr. pod red. V. A. Voznesenskogo Sovremenny’e metody’ optimizacii compozicionny’x materialov. Kyiv, Budivelnyk, 1983, 144 p.
Noviczkij P. V., Zograf I. A. Izdanie 2, pererabotannoe i dopolnennoe Ocenka pogreshnostej rezultatov izmerenij. Leningrad, Energoatomizdat, 1991, 304 p.
Ajvazyan S. А., Enyukov I. S., Meshalkin L. D., pod red. S. A. Aivazyan. Prikladnaya statistika: Issledovanie zavisimostej: Spravochnoe izdanie. Moscow, Finansy i statistika, 1985, 487 p.
Ajvazyan S. А., Mxitaryan V. S. Pikladnaya statistika. Osnovy ekonometriki: Uchebnik dlya vuzov, v 2 tomax, izdanie 2, ispravlennoe. Vol. 1, Teoriya veroyatnostej i prikladnaya statistika. Moscow, YUNITI-DANA, 2001, 656 p.
Malov S .V. Regressionny’j analiz. Teoreticheskie osnovy i prakticheskie rekomendacii. St.Petersbourg, Izdatelstvo SPbGU, 2013, 280 p.
Hinkelmann K. Kempthorne O. Design and Analysis of Experiments, Introduction to Experimental Design, 2 edition, Vol. 1,Wiley-Interscience, 2007, 631 p. (Wiley Series in Probability and Statistics).
Radchenko S. G. Formalizovanny’e i e’vristicheskie resheniya v regressionnom analize: monografiya. Kyiv, «Kornijchuk», 2015, 236 p.
Konovalova I. O., Berkovich Yu. A., Eroxin A. N. i dr. Optimizaciya svetodiodnoj sistemy’ osveshheniya vitaminnoj kosmicheskoj oranzherei, Aviakosmicheskaya i e’kologicheskaya mediсina, 2016, Vol. 50, No. 3, pp. 17–22.
Radchenko S. G. Metodologiya regressionnogo analiza : monografiya. Kyiv, «Kornijchuk», 2011, 376 p.
Radchenko S. G. Аnаliz metodov modelirovaniya slozhnih sistem, Matematichni mashini i sistemi, 2015, No. 4, pp. 123–127.
Laboratory of experimental statistical method of investigation (LESMI) [Electronic resource]. Access mode: http://www.n-t.org/sp/lesmi
Site of the Department Mechanical Engineering, Mechanical Engineering Institute, National Technical University of Ukraine «Kyiv Polytechnic Institute» [Electronic resource]. Access mode: http://tm-mmi.kpi.ua/index.php/ru/1/publications/
How to Cite
Copyright (c) 2018 S. G. Radchenko
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Creative Commons Licensing Notifications in the Copyright Notices
The journal allows the authors to hold the copyright without restrictions and to retain publishing rights without restrictions.
The journal allows readers to read, download, copy, distribute, print, search, or link to the full texts of its articles.
The journal allows to reuse and remixing of its content, in accordance with a Creative Commons license СС BY -SA.
Authors who publish with this journal agree to the following terms:
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License CC BY-SA that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.