UDC 621.382
DETERMINING SAMPLE SIZE TO RESEARCH IGNITION DELAY TIME
A. N. Shesterkin, Dr. Sc. (Tech.), full professor, Department of computing and applied mathematics, RSREU, Ryazan, Russia;
orcid.org/0000-0002-0633-0591, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Temporal processes in gas-discharge matrix indicators, both under autonomous excitation of elements and under ionization conditions, are quite accurately described by exponential distribution. The aim of this work is to analyze methods for calculating the number of elements in a sample that provide determination of ignition delay times statistics. Methods for calculating a sample size for estimating mathematical expectation of delay time, the number of measurements and the quantity of histogram intervals, which ensure that experimental distribution agrees with exponential parametric and nonparametric methods, and that the histogram is presented with necessary accuracy, are considered. Confidence intervals of mathematical expectation estimates, displacements and probabilities of random values falling into the histogram intervals are defined. Recommendations to determine sample size needed to build a histogram and to estimate mathematical expectation are formulated. The accuracy of the results was confirmed by statistical modeling.
Key words: gas discharge indicator, delay time of discharge occurrence, exponential distribution, estimation of average delay time, sample size, errors of statistical estimates, histogram, confidence interval, statistical modeling.