Method of Statistical Data Processing for Two-Stage Fatigue Tests

Abstract

Prevention of degradation processes and prediction of lifetime of aircraft structure is essential for keeping high-reliability level and aviation safety. Research has shown that it is necessary to have significant statistics of the tested specimens to describe such a complicated process as metal fatigue. This paper aims to process statistical data for two-stage fatigue tests. Based on a review of the literature and experimental statistics, analytical models of the rate of degradation processes were highlighted. On this basis, it is recommended to study the evolutionary process of fatigue behaviour. Two-stage fatigue tests of technically pure copper were carried out in the statistical aspect. Above 200 specimens were tested to obtain an nl-n2 diagram. Oscillation of physical-mechanical characteristics the metal has occurred that it leads to the data significant scattering. Probability density functions f(n1, n2) for experiment results was obtained. Further research is needed for probability distribution fitting and more accurate selection of bivariate distribution.