Discussion on uncertainty of dynamic balance param

  • Detail

Discussion on the uncertainty of dynamic balance parameters

at present, there are some nonstandard parameters in quantitatively giving the reliability of measuring instruments, and there are some misunderstandings and too complex in the uncertainty evaluation. This paper puts forward some views for discussion in combination with the international normative documents in recent years

1. Measurement accuracy and measurement instrument accuracy are only qualitative concepts

China's measurement technical specification JJF general measurement terms and definitions emphasizes their definitions as a qualitative concept according to the current international published international metrology general terms (VIM). This view is still maintained in the third revision of vim published by the Joint Committee on international metrology guidelines (jcgm) last year, and it is pointed out that it cannot be given by a single value

that is, the accuracy cannot be expressed as, for example, 0.4%; 16mg; ≤16mg; ± 16mg, etc

the index of measuring instrument shall be in accordance with jjf1001

2. Measurement uncertainty and measurement instrument uncertainty

the definition of measurement uncertainty is based on the guidelines for the expression of measurement uncertainty (GUM) published by ISO and other 7 international organizations in 1995 and China's measurement technical specification JJF evaluation and expression of measurement uncertainty as:

represents the dispersion reasonably given to the measured value and the parameters related to the measurement results

can also be defined as:

the measurement of the possible error of the measured estimated value given by the measurement results, or the evaluation of the range of the measured true value

therefore, in addition to more than 20000 trade employees, it is:

a) a quantitative expression of the suspicious degree of the measurement results

b) is only related to the measurement results

c) comes from various operation links to obtain measurement results. The "uncertainty of measuring instruments" appeared in the third revision of vim in jcgm04 for the first time in a formal term and was defined as: the uncertainty component caused by measuring instruments determined by the calibration of measuring instruments

it is a component of the uncertainty of the measured results obtained by the measuring instrument, which is often a very important one. If the supplier of the instrument provides it, it will be much simpler and more reliable for the user to evaluate the uncertainty of the measurement results. Unfortunately, at present, almost no manufacturer has evaluated and given this parameter

3. Some simplified ways in the evaluation of measurement uncertainty

a) ignore those components that do not affect the combined uncertainty by 0.1

b) try to use the first test results or uncertainty evaluation results of previous similar tests

c) the uncertainty of the test method given in the technical specification shall be used as much as possible to ensure the accuracy of the notch size, such as the repeatability, repeatability limit, reproducibility limit (reproducibility limit) of the method, etc

d) when it is not necessary to evaluate the expanded uncertainty up, only u can be given instead of evaluating the degrees of freedom of each component

e) only when the effective degree of freedom of the combined uncertainty is very small, such as less than 6, it is necessary to use the insertion method to calculate the inclusion factor KP

f) when the estimated value of the input quantity is likely to be correlated, the correlation coefficient r can be taken as + 1 or -1 in the case of strong correlation, and + 0.5 or -0.5 in the case of weak correlation

g) when the users who need to buy high and low temperature impact testing machine hope to find a reliable high and low temperature impact testing machine, the functional relationship between the input and output of the manufacturer is nonlinear, the relative standard uncertainty can be used to synthesize without calculating the partial derivative

h) in many cases, the components of uncertainty can be evaluated separately according to systematic effect and random effect and then combined to simplify the evaluation process

i) if a large degree of freedom is required, it is generally not less than 10, not too large. (end)

Copyright © 2011 JIN SHI