CC BY
64
https://doi.org/10.29013/ESR-19-9.10-45-47
Murodov Jurabek Norpulatovich, independent researcher, Tashkent state Technical University E-mail: [email protected]
QUANTIFICATION OF THE UNCERTAINTY OF MEASUREMENT RESULTS
Abstract. The article discusses the main stages of development of uncertainty and types of uncertainty. It also shows the quantitative characteristics of the results of uncertainty in the concepts of uncertainty.
Keywords: measurement uncertainty, measurement result, sources of uncertainty, reference value, measuring instruments, expanded uncertainty.
In the late sixties of the twentieth century, among ror? If this is a deviation from the truth, then what
specialists associated with measurements and interpretation of their results, a new concept arose - measurement uncertainty.
The system for assessing the quality of measurements that existed up to that time was based on the concept of "true value" of a measured quantity. However, metrology, as a science, has never operated with a similar concept. As a source, the "real value" of the measured value was taken. Thus, not knowing the true value of the quantity, metrologists allowed themselves to argue that the deviation from it (error) is equal to a certain number, and it does not matter how this deviation was found - theoretically or experimentally. How true this statement is is unknown!
In real life, metrologists and practitioners who are not related to standards have no such questions. For them, the true (or real) value was the value reproduced by a measuring instrument of higher accuracy.
And suddenly there appears not so new, but with a completely different meaning, the concept of "uncertainty". What it is?! How necessary is this concept and how is it related to those provisions of metrology as a science with which one is already accustomed to operate. Nowadays, no "suddenly" can arise. There were, and they could not be, objective reasons that made us think about the question: "What is the er-
is the truth? "The true meaning of the measured quantity, like truth in a broad, philosophical sense, is not given. Neither materialistic nor idealistic philosophies allow themselves assertions about a complete knowledge of the essence of things, including the "true values" of measured quantities. That is, using the concept of "measurement error", they usually operate with values that include not only a deviation from a certain value of a value conventionally accepted as true (real), but also with a number of Unknowns, namely with a capital letter Unknown, of the value, within which the "true value" of the measured quantity can be.
When presenting the measurement result, it is necessary to quantify its quality so that the person to whom this measurement result is intended can evaluate its reliability. Without this, it is impossible to compare the results of measurements with each other or to compare them with the norms indicated in regulatory documents. Therefore, it is necessary to have an easy-to-use, accessible to understanding and universally recognized methodology for characterizing the quality of the measurement result, that is, for assessing and expressing its "uncertainty".
The concept of uncertainty as a characteristic of the quality of the measurement result is relatively
Section 6. Technical sciences
new. The traditional, long-used in metrology terms are "error" and "error analysis". It is now generally accepted that even after all known or suspected error components have been evaluated and corresponding corrections have been made to the measurement result, there is still doubt about how accurately the measurement result represents the value of the measured quantity.
As already noted, in the context of market globalization, the task of creating a single method for assessing and expressing uncertainty is becoming more and more relevant, so that the results of measurements carried out in different countries can be easily compared with each other. In this case, the method should have been universal, i.e., applicable to all types of measurements and to all types of input data used in measurements, and the value directly used to express uncertainty should be internally consistent (should be directly derived from the components that make it up, and also be independent of how these components are grouped and from dividing the components into subcomponents) and allowing transmission (there should be the possibility of direct use of uncertainty about enennoy one result as component uncertainties in the evaluation of another dimension, wherein the first result) is used.
In many cases - in industry, trade, healthcare, security - it is necessary to present the measurement result with an indication of the interval within which, it can be assumed, is the majority of the distribution of values that can reasonably be attributed to the value to be measured.
Consequently, the method for assessing and expressing measurement uncertainty should be able to indicate such an interval, in particular, an interval, probability of coverage or level of confidence that actually matches the required one.
Based on the above assumptions, in 1978, recognizing the lack of international unity on the issue of expressing measurement uncertainty, the highest world authority in the field of metrology, the International Committee ofWeights and Measures (CIPM),
requested the International Bureau of Weights and Measures (BIPM) to consider this a problem. As a result of complex painstaking work, to which the national metrological laboratories of 32 countries were involved, the authoritative international organizations BIPM, ISO, IEC, OIML, IUPAC, IUPAP and IFCS developed in 1993 a "Guide for the expression of measurement uncertainty" (hereinafter referred to as the Guide).
To correctly understand issues related to the assessment and expression of uncertainty, we consider a number of terms given and used in the Guide:
- measurement uncertainty - a parameter associated with the measurement result, which characterizes the variance of values that could reasonably be attributed to the measured quantity. The parameter may be, for example, standard deviation.
- standard deviation - mean square error (or standard deviation of standard deviation) of the measurement result (or arithmetic mean);
- standard uncertainty - the uncertainty of the measurement result, expressed as a standard deviation;
- estimation of (uncertainties) by type A - method for estimating uncertainty by statistical analysis of a number of observations;
- estimation of (uncertainties) by type B - a method of estimating uncertainty in a different way than the statistical analysis of series of observations;
- total standard uncertainty - the standard uncertainty of the measurement result, when the result is obtained from the values of a number of other quantities, equal to the positive square root of the sum of the terms, the terms being the variances or covariances of these other quantities, weighted according to how the measurement result changes depending on the change these quantities;
- expanded uncertainty - a value that determines the interval around the measurement result, within which, as you might expect, there is a large part of the distribution of values that could reasonably be attributed to the measured value.
References:
1. Zakharov I. P. Uncertainty of measurements for dummies and ... superiors: I study / I. P. Zakharov, -Kharkov: 2013. - 36 p.
2. Zakharov I. P., Kukush V. D. Theory of uncertainty in measurements. Textbook: - Kharkov, Consum, 2002.- 256 p.
3. RMG 91-2009. State system for ensuring the uniformity of measurements. The joint use of the concepts of "measurement error" and "measurement uncertainty".
4. COOMET R/GM/21:2011. Use of the concepts "measurement uncertainty" and "measurement uncertainty". General principles (use of concepts "error of measurement" and "uncertainty of measurement". General principles (Approved at the 21st meeting of the COOMET Committee (April 27-28, 2011, Yerevan, Armenia).
