Estimation of long-term stability of precision inductance measures using different methods

Inductance is a physical quantity that characterizes the ability of a conductor to accumulate magnetic field energy when an electric current flows. Considering the small number of standards of a similar level of accuracy in the world, it is interesting to study the behavior of individual characteristics of such standards using known research methods. The long-term stability of measures is an important metrological characteristic that characterizes the change in the nominal value of a standard or measure over time. Regular metrological research of the long-term stability of precision inductance measures with nominal values of 10 mH and 100 mH have been carried out since 2009. Analysis of the long-term behavior of precision inductance measures included checking the primary measurement data for randomness and outliers using Abbe's and Grubb's tests. The detection of anomalous results significantly affects the estimation of the random error component and the standard uncertainty of the reproduction of the inductance measure. There is a significant change in the inductance of both measures over the entire observation period, which is confirmed by Abbe's test. At the same time, this change is most significant in the first years, and then the change in inductance occurs slowly. The analysis of the functional change in inductance using the Grubb's test showed that, depending on the research object itself, different analytical dependencies can be used for describing the functional trend. The criterion for choosing the best approximation can be both the 'criterion of the reliability of the approximation' and the standard deviation of the 'cleaned' measured data. The practical value of this work is the possibility of accumulating and integrating the obtained experimental data for standards of a similar accuracy class in different countries, which makes it possible to further improve the methods of their construction, stabilization of reproducible characteristics and operational features.