Mathematical and software for determining the error in modeling a measuring instrument

In the tasks of metrological synthesis, the task of determining the metrological characteristics of the measuring instrument is set. In modeling, the measuring instrument can be represented as a set of nodes whose parameters affect the measurement result. The case of determining the probability density of the total error of the measurement result for sequentially connected units of the measuring instrument is considered. The identification of the distribution law of the total error is carried out on the basis of a machine experiment. As an example, it is proposed to consider a case combining a constant value of an input quantity for which the error is defined as additive noise composed of independent quantities. A machine experiment is performed to iteratively search for the composition of the distribution laws of random independent quantities of neighboring nodes of the measuring instrument, the result of the composition is compared with the known distribution laws. Two cases of attribution of the law of distribution of the total random variable to the normal law or the law of arbitrary form are indicated. The estimation of the error of the measuring instrument is based on the calculation of probabilistic characteristics based on the found probability distribution density, which makes it possible to use a priori information about each of the nodes of the measuring instrument in the evaluation. It is proposed to consider mathematical expectation, variance and interval probability as characteristics of the accuracy of the identified density of the error distribution of the measurement result.

1. Novitsky P.V., Zograf I.A. Estimation of measurement results errors. 2nd ed., reprint. and add. Energoatomizdat; 1991. 303 p. (In Russ.)

2. Wentzel E.S. Probability theory. Textbook for higher technical educational institutions. M.:Nauka Publishing House: The main editorial office of physical and mathematical literature; 1969. 576 p. (In Russ.)

3. Johnson N., Lyon F. Statistics and experiment planning in engineering and science. Methods of data processing. Trans. from English. М.: Publishing house "Mir"; 1980. 510 p. (In Russ.)

4. Gnedenko B.V. Course of probability theory. 12nd ed., reprint. and add. M.: Editorial URSS; 2019. 456 p. (In Russ.)

5. Glazebnyy K.I., Romantsova N.V., Sokolov A.N. Algorithmic Support for Calculating the Compositions of Distribution Laws. 2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus). 2021:360–363. DOI: 10.1109/ElConRus51938.2021.9396547.

6. Iglin S.P. Mathematical calculations based on MATLAB. SPb.:Publishing house BHV; 2005. 640 p. (In Russ.)

7. Alekseev V.V., Dolidze R.V., Nedosekin D.D., ChernyavskyE.A. Workshop on probabilistic methods in measuring technology: Textbook for universities. SPb.: Publishing house Enegoatomizdat. St. Petersburg Department; 1993. 264 p. (In Russ.)

8. Tsvetkov E.I. Metrology. Models. Metrological analysis. Metrological synthesis. SPb.: Publishing House of SPbSETU "LETI"; 2014. 293 p. (In Russ.)

9. Iglin S.P. Probability theory and mathematical statistics based on MATLAB. Kharkov, Ukraine: Publishing house of NTU "KhPI"; 2006. 612 p. (In Russ.)

10. Tsvetkov E.I., Suloeva E.S. Analysis of the parameters that determine the reliability of the results of a verification of measuring instruments. Measurement Techniques. 2018;61(9):872–877. DOI: 10.1007/s11018-018-1517-z.