CONFIDENCE LIMITS OF MEASUREMENT RESULTS
Abstract
Problems relating to density distribution of estimated probability are solved on the basis of existing theoretical and methodological approaches to the calculation of confidence limits of measurement error. Attention is paid to the diversity of actual occurring distributions, including normal, trapezoidal, exponential Laplace distribution, the t-distribution, and so on. Particular emphasis of the authors is placed on the use of a method based on the simulation of arrays of numbers with a given distribution in the mathematical apparatus for calculating the confidence level. The simulation results with comments are given in seven tables that allow comparing values and justifying advantages and disadvantages of each type of distribution using the criterion of confidence in end values of measurements.
About the Authors
N. A. Rubichev
Moscow State University of Railway Engineering (MIIT), Moscow
Russian Federation
Russian Federation
Ph.D. (Tech), associate professor at the department of Power engineering of transport
N. B. Seydakhmetov
Moscow State University of Railway Engineering (MIIT), Moscow
Russian Federation
fifth-year student
Russian Federation
fifth-year student
References
1. Грановская В. А., Сирая Т. Н. Методы обработки экспериментальных данных при измерении. – Л.: Энергоатомиздат, 1990. – 288 с. Granovskaya, V.A., Siraya, T. N. Methods of experimental data processing in the measurement [Metody obrabotki eksperimental’nyh dannyh pri izmerenii]. Leningrad, Energoatomizdat publ., 1990, 288 p.
2. Фрумкин В. Д., Рубичев Н. А. Теория вероятностей и статистика в метрологии и измерительной технике. – М.: Машиностроение, 1987. – 168 с. Frumkin, V.D., Rubichev, N. A. Probability theory and statistics in metrology and measurement techniques [Teorija verojatnostej i statistika v metrologii i izmeritel’noj tehnike]. Moscow, Mashinostroenie publ., 1987, 168 p.
3. ГОСТ 8.207–76. ГСИ. Прямые измерения с многократными наблюдениями. Методы обработки результатов наблюдений. State standard 8.207-76. State System for Ensuring Uniform Measurement. Direct measurements with multiple observations. Methods for processing the results of observations [GOST 8.207-76. GSI. Prjamye izmerenija s mnogokratnymi nabljudenijami. Metody obrabotki rezul’tatov nabljudenij].
4. Руководство по выражению неопределенности измерения: Пер. с англ. – СПб.: ВНИИМ, 1999. – 134 с. Guide to the Expression of Uncertainty in Measurement [Rukovodstvo po vyrazheniju neopredelennosti izmerenija: Per. s angl]. St. Petersburg, VNIIM publ., 1999, 134 p.
5. Рубичев Н. А., Рябцев Г. Г. Типовые ошибки применения статистических методов обработки измерительной информации и способы их устранения // Метрология. – 2012. – № 6. – С. 3–16. Rubichev, N.A., Ryabtsev, G. G. Typical errors of applying statistical methods for processing measuring data and their solutions [Tipovye oshibki primenenija statisticheskih metodov obrabotki izmeritel’noj informacii i sposoby ih ustranenija]. Metrologija, 2012, No. 6, pp. 3–16.
Review
For citations:
Rubichev N.A., Seydakhmetov N.B. CONFIDENCE LIMITS OF MEASUREMENT RESULTS. World of Transport and Transportation. 2014;(5):38-51.
Views:
265
Email this article 