Calculation of the capacitance predicted value of a precision capacitor and evaluation of its uncertainty by extrapolating the results of previous calibrations using the weighted linear regression method

Obtaining the predicted value of the standard measure is a key element in ensuring the traceability and reliability of high-precision measurement results. In case, the intervals between calibrations are significant, and the drift of the standard can affect the accuracy of unit reproduction, there is a need to build a statistical model that allows extrapolating the value of the measure to a certain point in time. This approach provides the ability to maintain metrological reliability even in the absence of actual calibrations, which is especially important for national standards and comparisons between national metrological institutes. The article is devoted to the development of a method for calculating the predicted value of the quantity reproduced by the standard, taking into account the time since the last calibration, the uncertainty of drift extrapolation, the possible nonlinearity of the drift model, and the results of previous calibrations. The application of the method will allow users to perform measurements with a guaranteed known metrological result. This increases the consistency of measurements at the international level and contributes to the development of a mutual recognition system of calibration results, which is fundamental to modern metrology. The research materials are recommended for use by calibration, testing, and measurement laboratories, as well as national metrological institutes, during international comparisons of national standards.