On-line Determination of the Measurement Uncertainty of the Stochastic Measurement Method

− The paper presents a stochastic sampling method applied to a measurement of the mean value of the product of two arbitrary signals. This method is especially convenient for measuring electric power, electric energy consumption, as well as root-mean-square of voltage and/or current. The method features a very simple hardware and the possibility of selecting the measurement accuracy by choosing a suitable measurement interval. When measuring electric energy, accuracy better than 0.01 % is achieved. The measurement uncertainty of the stochastic sampling method is analysed and the influence of dominant factors is quantitatively derived. Special consideration is given to the influence of dispersion of A/D converter outputs and the waveforms of the two input signals as contributors to the measurement uncertainty. As the exact shapes of the waveforms are not known in advance, it is not possible to determine their influence on the measurement uncertainty. To overcome this issue, a novel additional hardware is designed, which provides the estimate of measurement uncertainty during the experiment simultaneously with the measurement result, its measurement uncertainty is available on-line.