|Generate QR code
Open Access
Subscription Access
Comparison of interval analysis methods and standard statistical ones in a problem of estimating experimental data with uncertainties
https://doi.org/10.32446/0368-1025it.2019-2-13-17
Abstract
The paper considers application of the interval analysis procedures to estimation of parameters of an experimental chemical process under conditions of corruption, uncertainty of errors probability characteristics, and short sample of measurements. For these reasons, application of the standard statistical approach is hampered; especially, it becomes impossible to determine accurately the confidential intervals for parameters of the process. Under such conditions, namely methods of the interval analysis can work reliably giving exact set of the admissible values of the parameters to be estimated. Such a set is very necessary for researches for proper organization of corresponding technological processes by correct choice of the parameters. In the investigation, existing interval analysis procedures were advanced and arranged for processing a concrete experimental data. As a comparison, approximate estimations of parameters have been calculated by the standard statistical approach. It is shown that the standard approach gives practically senseless estimations of the process parameters.
About the Authors
S. I. Kumkov
N.N. Krasovskii Institute of Mathematics and Mechanics of Ural Branch of Russian Academy of Sciences; Ural Federal University
Russian Federation
Russian Federation
Jaulin Luc
ENSTA-Bretagne
Russian Federation
Russian Federation
References
1. Вакарин С. В., Меляева А. А., Семерикова О. Л., Кондратюк В. С., Панкратов А. А., Плаксин С. В., Поротникова Н. М., Зауков Ю. П., Петров Д. А., Микушина Ю. В., Шишмаков А. Б., Чупахин О. Н. Каталитичекая активность оксида наноразмерных вольфрамовых бронз, получаемых электролизом расплавленных солей // Известия РАН. Сер. Химическая. 2011. № 10. С. 1051-1054.
2. R. B. Kearfott, M. Nakao, A. Neumaier, S. Rump, S. P. Shary, P. van Hentenryck. Standardized notation in interval analysis // J. Comput. Technol. 2010. V. 15. No. 1. P. 7-13.
3. Шарый С. П. Конечномерный интервальный анализ. [Электрон. книга] URL: http://www.nsc.ru/interval/Library/InteBooks (дата обращения 10.10.2018).
4. ГОСТ Р 40.2.028-2003. ГСИ. Рекомендации построения калибровочных характеристик. Оценивание погрешностей (неопределенностей) линейных калибровочных характеристик на основе метода наименьших квадратов.
5. Канторович Л. В. О новых подходах к вычислительным методам и обработке наблюдений // Сибирский математический журнал. 1962. Т. III. № 5. С. 701-709.
6. Jaulin L., Kieffer M., Didrit O., Walter E. Applied interval analysis. London: Springer-Verlag, 2001.
7. Hansen E., Walster G.W. Global optimization using interval analysis. New York, USA: Marcel Dekker Inc., 2004.
8. Kumkov S. I., Mikushina Yu. V. Interval approach to identification of catalytic process parameters // Reliable Computing. 2013. V. 19. Iss. 2. P. 197-214.
9. Kumkov S. I. Interval approach to robust bounded-error estimation of corrupted experimental data // SYSID-2012: Prоs. 16th IFAC Symposium on System Identification. July 11-13, 2012, Belgium, Brussels, Vrije Universiteit Brussels, ID0316. Р.1091-1096. [Электрон. версия] URL: http://IFAC.SYSID-2012 (дата обращения 10.10.2018).
10. Redkin A. A., Zaikov Yu. P., Kumkov S. I., et al. Heat Capacity of Molten Halides // J. Phys. Chem. B 2015, 119, 509-512. DOI: 10.1021/jp509932e.
11. Kumkov S. I. Applied techniques of interval analysis for estimation of experimental data // SCAN 2014: Book of Abstracts of 16th GAMM-IMACS Intern. Symp. Sci. Computing, Computer Arithmetic and Validated Numerics, September 21-26, 2014, Germany, Wurzburg, Wurzburg University. P. 90-91.
12. Архипов П. А., Кумков С. И. и др. Оценивание активности свинца в двойных системах Pb-Sb и Pb-Bi // Расплавы. 2012. № 5. С. 43-52.
13. Кумков С. И. Обработка экспериментальных данных ионной проводимости расплавленного электролита методами интервального анализа // Расплавы. 2010. № 3. С. 86-96.
14. Kumkov S. I., Kreinovich V., Pownuk A. In System Identification, Interval (and Fuzzy). Estimates Can Lead to Much Better Accuracy than the Traditional Statistical Ones: General Algorithm and Case Study // IEEE Intern. Conf. Systems, Man, and Cybernetics (SMC), October 5-8, 2017, Canada, Banff, Banff Center.
Review
For citations:
Kumkov S.I., Luc J. Comparison of interval analysis methods and standard statistical ones in a problem of estimating experimental data with uncertainties. Izmeritel`naya Tekhnika. 2019;(2):13-17. (In Russ.) https://doi.org/10.32446/0368-1025it.2019-2-13-17
Views:
97
Email this article 