The results of statistical measurements of the resistance of the metal film resistor are given below. The analysis of the dependencies of the standard deviation on the number of measurements σR(N)envisages a tendency of raising σR with the number of measurements. It indicates that the obtained results are not independent since fluctuations of them are not of “white” noise type. The dependence of the standard deviation of measurement results onN confirms the results of other studies.
A hypothesis about the impact of the non-equilibrium state of the studiedobject on the dispersion of measurement results is proposed. The energy spectrum of the measurement results is uneven and similar to flicker noise’s one, as the object under study is in non-equilibrium state, which affect the dispersion of results.
The dependence of the normalized standard deviation of the measurement results on the number of measurements is given.It is compared the normalized standard deviations: under the condition that S(f) has the form of “white” noise (an equilibrium system) and under the condition of combination of “white” noise and flicker noise. From the above dependence it can be seen that averaging 100 measurements gives a 10-fold decrease in the standard deviation of the measured value. If the form of “white” noise is inherent in the energy spectrum S(f), then a decrement of standard deviation in ~1.4 times is noticed. For energy spectrum of flicker noise the decrement is ≈ 1.7 times.
Real measured spectrum is similar to flicker noise’s one. So, the random error of measurement cannot be reduced to an arbitrarily small value by averaging a large number of measurement results. Only if the Tmeasmax ≈ Mmaxτ ratio is considered, where Tmeasmax is the time of measurement of the fluctuating parameters of the real investigated system, τ is the relaxation time of the investigated system, Mmax is the number of possible ways in which the equilibrium state of the system can be implemented, therandom error approximates to the minimum value.
[1] I. Gorban, "Evaluation of statistically unpredictable changes in physical quantities over large observation intervals", Journ. Techn. Phys., vol. 88, iss. 12, p. 1779-1786, 2018.
https://doi.org/10.1134/S106378421812006X
[2] I. Gorban, The phenomenon of statistical stability, Kyiv, Ukraine: Scientific thought, 2014.
[3] I. Gorban, The statistical stability phenomenon, Springer , 2017.
https://doi.org/10.1007/978-3-319-43585-5
[4] I. Gorban, Randomness and hyper-randomness, Kyiv, Ukraine: Scientific thought, 2016.
[5] I. Gorban, Ramdomness and hyper-randomness. Springer , 2 0 18 .
[6] V. Eskov, T. Gavrilenko, V. Eskov and al., "The phenomenon of statistical instability of systems of the third type-complexity", Journ. Techn. Phys., vol. 87, iss. 11, p. 1609-1614, 2017.
https://doi.org/10.1134/S106378421711007X
[7] I. Gorban, "The phenomenon of statistical stability", Journ. Techn. Phys., vol . 59, iss. 3, p. 22-30, 2014.
https://doi.org/10.1134/S1063784214030128
[8] H. Nyquist, "Thermal agitation of electric charge in conductors", Phys. Rev., vol. 32, no. 1, p. 110-113, 1928.
https://doi.org/10.1103/PhysRev.32.110
[9] Z. Kolodiy, B. Stadnyk, S. Yatsyshyn, "Development of Noise Measurements. Part 2. Random Error", Sensors & Transducers, vol. 151, iss. 4, p. 107-112, 2013.
[10] Z. Kolodiy, S. Yatsyshyn, "Entropy of Measurements of Electric and Non-electric Systems Fluctuating Parameters", Journ. Electr. Res. & Appl., vol. 2, iss. 3, p. 26-30, 2018.
https://doi.org/10.26689/jera.v2i3.383